PRINCIPLES OF HEALTH ECONOMICS

for non-economists

Xavier Martinez-Giralt

January 2008
This version March 2010

CODE
Center for the study of Organizations
and Decisions in Economics
 Copyright c 2008 Xavier Martinez-Giralt.
Permission is granted to copy, distribute and/or modify this
document under the terms of the GNU Free Documentation
License, Version 1.2 or any later version published by the Free
Software Foundation; with the Invariant Sections, with the
Front-Cover Texts, and with the Back-Cover Texts. A copy of
the license is included in the section entitled “GNU Free
Documentation License”.
 HEALTH ECONOMICS

1. Economics and Health Economics

1.1 What is economics about?
1.2 What is health economics? Elements of HE;
Organization, actors of the health care market;
Structure of a health care system
2. The agents of the economy
2.1 Demand: consumers, patients, elasticity
2.2 Supply: firms, hospitals physicians;
Efficiency, Efficacy, Effectiveness, Equity,
Opportunity cost
2.3 Insurers
3. The market and the health care market
3.1 Why is the health care market different?
3.2 Perfectly competitive markets
4. Regulation
4.1 The public sector
4.2 Mechanisms of regulation
4.3 Reasons for regulation
4.4 Regulation in the health care market
5. Public goods
6. Nonprofit organizations
6.1 Why do nonprofit enterprises exist?
6.2 Modeling a nonprofit hospital
7. A health policy exercise
 8. Uncertainty, risk and insurance
8.1 Attitudes facing risk
8.2 Health insurance
9. Contract theory
9.1 Contracts, information and agency relation
9.2 Adverse selection, moral hazard and signalling
9.3 Supplier induced demand
10. Economic evaluation
10.1 QALYs
10.2 Components in economic evaluation
10.3 CEA, CUA, CBA
11. Macroeconomics
11.1 What is macroeconomics about?
11.2 The working of the economy
11.3 Macroeconomics of the health sector
References

Deeming, C., and J. Keen, 2004, Choice and equity: lessons

from long term care, British Medical Journal, 328: 1390-
1391.
Drummond, M.F., B, O’Brien, G.L. Stoddart y G.W. Torrance,
2003, Methods for the economic evaluation of health care
programmes, Oxford University Press, Oxford
Feldstein, P.J., 2004, Health Care Economics, Thomson, Lon-
don.
Folland, S., A.C. Goodman, and M. Stano, 2009, The Eco-
nomics of Health and Health Care, Pearson Education,
Upper Saddle River, New Jersey.
Jacobs, P., 2003, The Economics of Health and Medical Care,
Jones and Bartlett Publishers.
Jones, A.M., 2006, The Elgar Companion to Health Economics,
Cheltenham, Edward Elgar.
Keating, B., 2008, Microeconomics for Public Managers, Black-
well, London.
Lewis, T.R., J.H. Reichman, and A.D. So, 2007, The case for
public funding and public oversight of clinical trials, Economists’
Voice, January.
Macho Stadler, I. and J. D. Pérez Castrillo, 2001, An Introduc-
tion to the Economics of Information, New York, Oxford
University Press.
McGuire, A., J. Henderson, and G. Mooney, 1995, The Eco-
nomics of Health Care, Routledge, London.
Phelps, C.E., 2009, Health Economics, Addison Wesley, New
York.
Porter, M.E., and E.O. Teisberg, 2006, Redefining Health Care,
Boston (Ma.), Harvard Business School.
Santerre, R.E., and S.P. Neun, 2004, Health Economics, Thom-
son, London.
Wonderling, D., R. Gruen, and N. Black, 2005, Introduction to
Health Economics, Open University Press, Maidenhead,
Berkshire.
Zweifel, P., F. Breyer, and M. Kifmann, 2009, Health Eco-
nomics, Springer, London.
 GNU Free Documentation License

Version 1.2, November 2002

Copyright 2000,2001,2002
c Free Software Foundation, Inc.
59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
Everyone is permitted to copy and distribute verbatim copies of this license
document, but changing it is not allowed.

Preamble
The purpose of this License is to make a manual, textbook, or other functional and
useful document ”free” in the sense of freedom: to assure everyone the effective
freedom to copy and redistribute it, with or without modifying it, either commer-
cially or noncommercially. Secondarily, this License preserves for the author and
publisher a way to get credit for their work, while not being considered responsible
for modifications made by others.
This License is a kind of ”copyleft”, which means that derivative works of the
document must themselves be free in the same sense. It complements the GNU
General Public License, which is a copyleft license designed for free software.
We have designed this License in order to use it for manuals for free software,
because free software needs free documentation: a free program should come with
manuals providing the same freedoms that the software does. But this License is
not limited to software manuals; it can be used for any textual work, regardless of
subject matter or whether it is published as a printed book. We recommend this
License principally for works whose purpose is instruction or reference.

1 Applicability and Definitions

This License applies to any manual or other work, in any medium, that contains a
notice placed by the copyright holder saying it can be distributed under the terms
of this License. Such a notice grants a world-wide, royalty-free license, unlimited
in duration, to use that work under the conditions stated herein. The ”Document”,
below, refers to any such manual or work. Any member of the public is a licensee,
and is addressed as ”you”. You accept the license if you copy, modify or distribute
the work in a way requiring permission under copyright law.
A ”Modified Version” of the Document means any work containing the Doc-
ument or a portion of it, either copied verbatim, or with modifications and/or trans-
lated into another language.
A ”Secondary Section” is a named appendix or a front-matter section of the
Document that deals exclusively with the relationship of the publishers or authors
of the Document to the Document’s overall subject (or to related matters) and con-
tains nothing that could fall directly within that overall subject. (Thus, if the Doc-
ument is in part a textbook of mathematics, a Secondary Section may not explain
any mathematics.) The relationship could be a matter of historical connection with
the subject or with related matters, or of legal, commercial, philosophical, ethical
or political position regarding them.
The ”Invariant Sections” are certain Secondary Sections whose titles are des-
ignated, as being those of Invariant Sections, in the notice that says that the Docu-
ment is released under this License. If a section does not fit the above definition of
Secondary then it is not allowed to be designated as Invariant. The Document may
contain zero Invariant Sections. If the Document does not identify any Invariant
Sections then there are none.
The ”Cover Texts” are certain short passages of text that are listed, as Front-
Cover Texts or Back-Cover Texts, in the notice that says that the Document is
released under this License. A Front-Cover Text may be at most 5 words, and a
Back-Cover Text may be at most 25 words.
A ”Transparent” copy of the Document means a machine-readable copy, rep-
resented in a format whose specification is available to the general public, that is
suitable for revising the document straightforwardly with generic text editors or
(for images composed of pixels) generic paint programs or (for drawings) some
widely available drawing editor, and that is suitable for input to text formatters or
for automatic translation to a variety of formats suitable for input to text formatters.
A copy made in an otherwise Transparent file format whose markup, or absence
of markup, has been arranged to thwart or discourage subsequent modification by
readers is not Transparent. An image format is not Transparent if used for any
substantial amount of text. A copy that is not ”Transparent” is called ”Opaque”.
Examples of suitable formats for Transparent copies include plain ASCII with-
out markup, Texinfo input format, LaTeX input format, SGML or XML using
a publicly available DTD, and standard-conforming simple HTML, PostScript or
PDF designed for human modification. Examples of transparent image formats
include PNG, XCF and JPG. Opaque formats include proprietary formats that can
be read and edited only by proprietary word processors, SGML or XML for which
the DTD and/or processing tools are not generally available, and the machine-
generated HTML, PostScript or PDF produced by some word processors for output
purposes only.
The ”Title Page” means, for a printed book, the title page itself, plus such
following pages as are needed to hold, legibly, the material this License requires
to appear in the title page. For works in formats which do not have any title page
as such, ”Title Page” means the text near the most prominent appearance of the
work’s title, preceding the beginning of the body of the text.
A section ”Entitled XYZ” means a named subunit of the Document whose title
either is precisely XYZ or contains XYZ in parentheses following text that trans-
lates XYZ in another language. (Here XYZ stands for a specific section name men-
tioned below, such as ”Acknowledgements”, ”Dedications”, ”Endorsements”,
or ”History”.) To ”Preserve the Title” of such a section when you modify the
Document means that it remains a section ”Entitled XYZ” according to this defini-
tion.
The Document may include Warranty Disclaimers next to the notice which
states that this License applies to the Document. These Warranty Disclaimers are
considered to be included by reference in this License, but only as regards dis-
claiming warranties: any other implication that these Warranty Disclaimers may
have is void and has no effect on the meaning of this License.

2 Verbatim Copying
You may copy and distribute the Document in any medium, either commercially or
noncommercially, provided that this License, the copyright notices, and the license
notice saying this License applies to the Document are reproduced in all copies,
and that you add no other conditions whatsoever to those of this License. You may
not use technical measures to obstruct or control the reading or further copying
of the copies you make or distribute. However, you may accept compensation in
exchange for copies. If you distribute a large enough number of copies you must
also follow the conditions in section 3.
You may also lend copies, under the same conditions stated above, and you
may publicly display copies.

3 Copying in Quantity
If you publish printed copies (or copies in media that commonly have printed cov-
ers) of the Document, numbering more than 100, and the Document’s license no-
tice requires Cover Texts, you must enclose the copies in covers that carry, clearly
and legibly, all these Cover Texts: Front-Cover Texts on the front cover, and Back-
Cover Texts on the back cover. Both covers must also clearly and legibly identify
you as the publisher of these copies. The front cover must present the full title with
all words of the title equally prominent and visible. You may add other material on
the covers in addition. Copying with changes limited to the covers, as long as they
preserve the title of the Document and satisfy these conditions, can be treated as
verbatim copying in other respects.
If the required texts for either cover are too voluminous to fit legibly, you
should put the first ones listed (as many as fit reasonably) on the actual cover,
and continue the rest onto adjacent pages.
If you publish or distribute Opaque copies of the Document numbering more
than 100, you must either include a machine-readable Transparent copy along with
each Opaque copy, or state in or with each Opaque copy a computer-network lo-
cation from which the general network-using public has access to download using
public-standard network protocols a complete Transparent copy of the Document,
free of added material. If you use the latter option, you must take reasonably pru-
dent steps, when you begin distribution of Opaque copies in quantity, to ensure that
this Transparent copy will remain thus accessible at the stated location until at least
one year after the last time you distribute an Opaque copy (directly or through your
agents or retailers) of that edition to the public.
It is requested, but not required, that you contact the authors of the Document
well before redistributing any large number of copies, to give them a chance to
provide you with an updated version of the Document.

4 Modifications
You may copy and distribute a Modified Version of the Document under the con-
ditions of sections 2 and 3 above, provided that you release the Modified Version
under precisely this License, with the Modified Version filling the role of the Docu-
ment, thus licensing distribution and modification of the Modified Version to who-
ever possesses a copy of it. In addition, you must do these things in the Modified
Version:

A. Use in the Title Page (and on the covers, if any) a title distinct from that of
the Document, and from those of previous versions (which should, if there
were any, be listed in the History section of the Document). You may use the
same title as a previous version if the original publisher of that version gives
permission.

B. List on the Title Page, as authors, one or more persons or entities respon-
sible for authorship of the modifications in the Modified Version, together
with at least five of the principal authors of the Document (all of its prin-
cipal authors, if it has fewer than five), unless they release you from this
requirement.

C. State on the Title page the name of the publisher of the Modified Version, as
the publisher.

D. Preserve all the copyright notices of the Document.

 E. Add an appropriate copyright notice for your modifications adjacent to the
other copyright notices.

F. Include, immediately after the copyright notices, a license notice giving the
public permission to use the Modified Version under the terms of this Li-
cense, in the form shown in the Addendum below.

G. Preserve in that license notice the full lists of Invariant Sections and required
Cover Texts given in the Document’s license notice.

H. Include an unaltered copy of this License.

I. Preserve the section Entitled ”History”, Preserve its Title, and add to it an
item stating at least the title, year, new authors, and publisher of the Modified
Version as given on the Title Page. If there is no section Entitled ”History”
in the Document, create one stating the title, year, authors, and publisher of
the Document as given on its Title Page, then add an item describing the
Modified Version as stated in the previous sentence.

J. Preserve the network location, if any, given in the Document for public ac-
cess to a Transparent copy of the Document, and likewise the network loca-
tions given in the Document for previous versions it was based on. These
may be placed in the ”History” section. You may omit a network location
for a work that was published at least four years before the Document itself,
or if the original publisher of the version it refers to gives permission.

K. For any section Entitled ”Acknowledgements” or ”Dedications”, Preserve

the Title of the section, and preserve in the section all the substance and
tone of each of the contributor acknowledgements and/or dedications given
therein.

L. Preserve all the Invariant Sections of the Document, unaltered in their text
and in their titles. Section numbers or the equivalent are not considered part
of the section titles.

M. Delete any section Entitled ”Endorsements”. Such a section may not be

included in the Modified Version.

N. Do not retitle any existing section to be Entitled ”Endorsements” or to con-

flict in title with any Invariant Section.

O. Preserve any Warranty Disclaimers.

If the Modified Version includes new front-matter sections or appendices that

qualify as Secondary Sections and contain no material copied from the Document,
you may at your option designate some or all of these sections as invariant. To
do this, add their titles to the list of Invariant Sections in the Modified Version’s
license notice. These titles must be distinct from any other section titles.
 You may add a section Entitled ”Endorsements”, provided it contains nothing
but endorsements of your Modified Version by various parties–for example, state-
ments of peer review or that the text has been approved by an organization as the
authoritative definition of a standard.
You may add a passage of up to five words as a Front-Cover Text, and a passage
of up to 25 words as a Back-Cover Text, to the end of the list of Cover Texts in the
Modified Version. Only one passage of Front-Cover Text and one of Back-Cover
Text may be added by (or through arrangements made by) any one entity. If the
Document already includes a cover text for the same cover, previously added by
you or by arrangement made by the same entity you are acting on behalf of, you
may not add another; but you may replace the old one, on explicit permission from
the previous publisher that added the old one.
The author(s) and publisher(s) of the Document do not by this License give
permission to use their names for publicity for or to assert or imply endorsement
of any Modified Version.

5 Combining Documents
You may combine the Document with other documents released under this License,
under the terms defined in section 4 above for modified versions, provided that
you include in the combination all of the Invariant Sections of all of the original
documents, unmodified, and list them all as Invariant Sections of your combined
work in its license notice, and that you preserve all their Warranty Disclaimers.
The combined work need only contain one copy of this License, and multiple
identical Invariant Sections may be replaced with a single copy. If there are mul-
tiple Invariant Sections with the same name but different contents, make the title
of each such section unique by adding at the end of it, in parentheses, the name of
the original author or publisher of that section if known, or else a unique number.
Make the same adjustment to the section titles in the list of Invariant Sections in
the license notice of the combined work.
In the combination, you must combine any sections Entitled ”History” in the
various original documents, forming one section Entitled ”History”; likewise com-
bine any sections Entitled ”Acknowledgements”, and any sections Entitled ”Dedi-
cations”. You must delete all sections Entitled ”Endorsements”.

6 Collections of Documents
You may make a collection consisting of the Document and other documents re-
leased under this License, and replace the individual copies of this License in the
various documents with a single copy that is included in the collection, provided
that you follow the rules of this License for verbatim copying of each of the docu-
ments in all other respects.
 You may extract a single document from such a collection, and distribute it
individually under this License, provided you insert a copy of this License into
the extracted document, and follow this License in all other respects regarding
verbatim copying of that document.

7 Aggregation With Independent Works

A compilation of the Document or its derivatives with other separate and indepen-
dent documents or works, in or on a volume of a storage or distribution medium, is
called an ”aggregate” if the copyright resulting from the compilation is not used to
limit the legal rights of the compilation’s users beyond what the individual works
permit. When the Document is included in an aggregate, this License does not ap-
ply to the other works in the aggregate which are not themselves derivative works
of the Document.
If the Cover Text requirement of section 3 is applicable to these copies of
the Document, then if the Document is less than one half of the entire aggregate,
the Document’s Cover Texts may be placed on covers that bracket the Document
within the aggregate, or the electronic equivalent of covers if the Document is in
electronic form. Otherwise they must appear on printed covers that bracket the
whole aggregate.

8 Translation
Translation is considered a kind of modification, so you may distribute translations
of the Document under the terms of section 4. Replacing Invariant Sections with
translations requires special permission from their copyright holders, but you may
include translations of some or all Invariant Sections in addition to the original ver-
sions of these Invariant Sections. You may include a translation of this License, and
all the license notices in the Document, and any Warranty Disclaimers, provided
that you also include the original English version of this License and the original
versions of those notices and disclaimers. In case of a disagreement between the
translation and the original version of this License or a notice or disclaimer, the
original version will prevail.
If a section in the Document is Entitled ”Acknowledgements”, ”Dedications”,
or ”History”, the requirement (section 4) to Preserve its Title (section 1) will typi-
cally require changing the actual title.

9 Termination
You may not copy, modify, sublicense, or distribute the Document except as ex-
pressly provided for under this License. Any other attempt to copy, modify, sub-
license or distribute the Document is void, and will automatically terminate your
rights under this License. However, parties who have received copies, or rights,
from you under this License will not have their licenses terminated so long as such
parties remain in full compliance.

10 Future Revisions of This License

The Free Software Foundation may publish new, revised versions of the GNU Free
Documentation License from time to time. Such new versions will be similar in
spirit to the present version, but may differ in detail to address new problems or
concerns. See http://www.gnu.org/copyleft/.
Each version of the License is given a distinguishing version number. If the
Document specifies that a particular numbered version of this License ”or any later
version” applies to it, you have the option of following the terms and conditions
either of that specified version or of any later version that has been published (not
as a draft) by the Free Software Foundation. If the Document does not specify a
version number of this License, you may choose any version ever published (not
as a draft) by the Free Software Foundation.
 HEALTH ECONOMICS

lt
ira
1. Economics and Health Economics

G
z-
ne
ti
Economics. What is this?

ar
M
er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

1
Economics: Study of the way in which economic
agents take their decisions regarding the use (al-
location) of scarce resources.

lt
ira
G
Economic agents: Decision makers in the economy.

z-
ne
Individuals, households, enterprises (for profit, non-

ti
ar
profit; production, distribution), State.

M
er
vi
Xa
Decisions:

c
s
ic
om

- what to produce/consume?
on
Ec

- how much to produce/consume?

lth
ea
H

- How to produce/consume?
of
es
pl
ci

- Who produces/consumes?
in
Pr

Answers to these questions depend on the organi-

zation of the economy: central plan, free market,
mixed systems.

1-a
 lt
ira
G
Reality too complex. Study of an economy by means

z-
ne
of models (theories): set of assumptions providing a

ti
ar
simplified representation of reality capturing the fun-

M
er
damental relationships among economic agents [→
vi
Xa
road map vs. road network].

c
s
ic
om

Two (complementary) uses of models:

on
Ec

- description of decision making process → positive

lth
ea

economics
H
of
es

- policy design (control and improvement of decision

pl
ci

making) → normative economics

in
Pr

1-b
 lt
ira
G
z-
ne
ti
ar
M
er
vi
Xa
Scarcity: wants vs. limited resources

c
s
ic
om

F Why do people demand (want) health care?

on
Ec

(i) Healthy status → 4 income → 4 leisure

lth
ea
H
of
es
pl
ci
in
Pr

1-c
 Pr
in
ci
pl
es
of
H
ea
lth
(ii) Population aging.

Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G

1-d
ira
lt
(iii) Increasing real income
- People with mild osteoarthritis of the knee often

lt
ira
have an operation than give up golf.

G
z-
ne
ti
- 4 income → 4 people’s expectations of health

ar
M
care: less prepared to put up pain, discomfort, lack

er
of mobility, ...
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

1-e
(iv) Improvement in medical technology:

lt
ira
G
- Technology increases range of possible treatments.

z-
ne
ti
ar
- Newer technology, more expensive

M
er
vi
Xa
e.g. kidney dialysis → prevent people dying from

c

kidney failure ⇒ machine is expensive, there are

s
ic
om

more patients (population aging) and they are treated

on

longer (extended life expectancy).

Ec
lth
ea

Reading:
H

Cutler, D., E. Glaeser, and A. Rosen, 2007, Is the

of
es

U.S. Population Behaving Healthier?, NBER Work-

pl
ci

ing Paper No. 13013

in
Pr

[http://www.nber.org/papers/w13013]

1-f
 Publications | Research | Data | About | People Login or change prefs

lt
ira
Is the U.S. Population Behaving Healthier?

G
" Reduced smoking, better control of medical risk factors such as hypertension and cholesterol, and better
education among the older population have been more important for mortality than the substantial increase in

z-
obesity."

ne
Knowing whether health behaviors are improving over time is important in forecasting medical
costs. And, a population that behaves in a healthier way will have a higher quality of life than one

ti
with a more adverse behavioral profile, even given length of life. In Is the U.S. Population
Behaving Healthier? (NBER Working Paper No. 13013), authors David Cutler, Edward Glaeser,

ar
and Allison Rosen consider what has happened to the population's health behaviors over time and
what the future may hold.

M
They find that the impact on longevity of trends in health behavior has not been uniform across

er
different behaviors over the past three decades. For example, while fewer people smoke than used
to, more people are obese. Examining these factors as a whole, the authors find significant

vi
improvement in the health-risk profile of the U.S. population between the early 1970s and the early
2000s. Reduced smoking, better control of medical risk factors such as hypertension and
Xa
cholesterol, and better education among the older population have been more important for
mortality than the substantial increase in obesity.

c

The results suggest substantial caution about the future, though. Where reductions in smoking can
be expected to have a continued impact on improved health, future changes in obesity might more
s

than overwhelm this trend. Two-thirds of the U.S. population is now overweight or obese. As a
ic

result, continued increases in weight from current levels will have a bigger impact on health than
did increases in weight from lower levels of Body Mass Index (BMI).
om

A large part of the impact of BMI is moderated through its effect on hypertension and high
cholesterol. Given that not everyone with these conditions takes medications, or is controlled by the
on

medication they do take, the resulting impact of rising weight on health can be significant. The
optimistic side of this picture, however, is the potential for better control of obesity. If the
Ec

effectiveness of risk-factor control can be increased, through more people taking medication and
those taking it using it more regularly, much of the impact of obesity on mortality risk can be
blunted, according to the authors.
lth

Understanding how to improve utilization of and adherence to recommended medications are key
issues in health outcomes. The research to date has focused on two possible avenues. The first is
ea

performance-based payment: physicians are now paid for office visits, but not for ensuring follow-
up with their recommendations. The idea behind pay-for-performance systems is to reward
H

physicians (or insurance companies) for successful efforts to increase utilization and possibly
adherence. Such efforts might involve having nurse outreach, automatic medication refills, or more
convenient office hours to monitor side effects.
of

The second strategy involves use of information technology. Patients can receive electronic
es

reminders about medication goals, information such as blood pressure can be transmitted and
monitored electronically, and automated decision tools can help with dosing and medication
pl

switches. Whether these or other strategies offer the greatest promise of improved adherence is
uncertain. The authors' results suggest that evaluating these strategies in practice is a high
ci

research priority.
in

The authors use as their primary data source the National Health and Nutrition Examination Survey
(NHANES). In the United States, it is the leading survey and includes both physical examination
Pr

and laboratory measurements. The authors use two NHANES surveys, the first from 1971-5
(NHANES I) and the second from 1999-2002 (NHANES IV). Their analysis begins with NHANES I
because it is the first population health survey that asked about smoking status, a key variable in
health risk.

-- Les Picker

The Digest is not copyrighted and may be reproduced freely with appropriate attribution of source.

1-g
F Resources: inputs, factors of production.

lt
ira
- land (physical resources of the planet)

G
z-
- labor (human resources)

ne
ti
- capital (resources created by human to aid in pro-

ar
M
duction: tools, machinery, factories, ...)

er
vi
Xa

enterprise: organization of resources to produce goods

c
s

and services.
ic
om
on
Ec

F Main concepts related with scarcity:

lth
ea

Efficiency
H
of
es
pl

Opportunity cost
ci
in
Pr

Production Possibility Frontier

1-h
What is Health Economics?

Allocation of resources within the health system in

the economy, as well as the functioning of the health

lt
care markets.

ira
G
z-
health system: set of interrelated elements (envi-

ne
ti
ronment, education, labor conditions, etc) having

ar
M
as objective the transformation of some sanitary re-

er
sources (inputs) into a health status (final output)
vi
Xa
through the production of health care services (in-

c

termediate output).
s
ic
om

Health vs. Health care:

on

Health is lack of illness → illness: restrictions im-

Ec

posed on the development of daily activities ⇒ value

lth
ea

in use but no value in exchange.

H

Health care: provision of services to improve health

of

status of individuals ⇒ intermediate output. Can be

es
pl

traded.
ci
in
Pr

healthcare market: interaction between providers and

consumers of health care services (and insurers).

Organization of HC market → crucial element of

analysis of HC system.
2
 lt
ira
Readings:

G
z-
ne
Rout, H.S., and P.K. Panda, 2007, Health and Health

ti
ar
Economics: A conceptual framework, in Health Eco-

M
er
nomics in India, edited by H.S. Rout, and P.K. Panda,
vi
Xa
New Century Publications, New Delhi: 13-29.

c
s
ic
om

[http://mpra.ub.uni-muenchen.de/6546]
on
Ec

Chicaiza, L, M. Garcı́a, and J. Lozano, 2008, Bring-

lth
ea

ing Institutions into Health Economics, Documentos

H

de Trabajo - Esc. de Admon. y Con Pub. 004687,

of
es

Universidad Nacional de Colombia - RCE.

pl
ci
in
Pr

[http://ideas.repec.org/p/col/000179/004687.html]

2-a
Health economics

- Descriptive studies: long tradition

lt
- Analytic studies: (relatively) recent. Stimulated by

ira
• (EU) Maastricht: 4 difficulties finance universal

G
z-
public health systems

ne
ti
• (US) Efforts extend coverage beyond Medicaid and

ar
M
Medicare - Clinton+Obama administration

er
⇓
vi
Xa
Restructure health care systems via

c

- stimulating competition
s
ic

- incentives: principals, agents, payment systems,

om

insurance, risk, etc.

on
Ec

Taking into account the characterisitics of the health

lth
ea

care system. (see p. 2n)

H
of

Consequences
es
pl
ci
in

Health economics as separate discipline from In-

Pr

dustrial economics:

- scientific journals

- huge volume of resources

2-b
Why has Health Economics developed into a disci-

lt
pline itself?

ira
G
z-
ne
Size and differential characteristics of health care

ti
ar
sector in the economy.

M
er
vi
Xa
Size

c
s
ic
om

OECD Health Data:

on
Ec

(a) Health expenditure

lth
ea
H

(b) Pharma expenditure

of
es
pl
ci

(c) Health financing

in
Pr

(d) Population

2-c
Total expenditure on health - % of gross domestic product

1960 1965 1970 1975 1980 1985 1990 1995 2000 2001 2002 2003 2004
Australia 4.0 4.2 4.5 6.9 6.8 7.2 7.5 8.0 8.8 8.9 9.1 9.2 ..
Austria 4.3 4.6 5.2 7.0 7.5 6.5 7.0 9.7 9.4 9.5 9.5 9.6 9.6
Belgium .. .. 3.9 5.6 6.3 7.0 7.2 8.2 8.6 8.7 8.9 10.1 ..
Canada 5.4 5.9 7.0 7.1 7.1 8.2 9.0 9.2 8.9 9.4 9.7 9.9 9.9
Czech Republic .. .. .. .. .. .. 4.7 7.0 6.7 7.0 7.2 7.5 7.3
Denmark .. .. 7.9 8.7 8.9 8.5 8.3 8.1 8.3 8.6 8.8 8.9 8.9
Finland 3.8 4.8 5.6 6.2 6.3 7.1 7.8 7.4 6.7 6.9 7.2 7.4 7.5
France 3.8 4.7 5.3 6.4 7.0 7.9 8.4 9.4 9.2 9.3 10.0 10.4 10.5

lt
Germany .. .. 6.2 8.6 8.7 9.0 8.5 10.3 10.4 10.6 10.8 10.9 ..

ira
Greece .. .. 6.1 6.6 7.4 7.4 9.6 9.9 10.4 10.3 10.5 10.0
Hungary .. .. .. .. .. .. 7.1 7.4 7.1 7.3 7.7 8.3 8.3

G
Iceland 3.0 3.5 4.7 5.7 6.2 7.2 7.9 8.4 9.2 9.3 10.0 10.5 10.2

z-
Ireland 3.7 4.0 5.1 7.3 8.3 7.5 6.1 6.7 6.3 6.8 7.2 7.2 7.1
Italy .. .. .. .. .. 7.5 7.7 7.1 7.9 8.0 8.2 8.2 8.4

ne
Japan 3.0 4.4 4.5 5.6 6.5 6.7 5.9 6.8 7.6 7.8 7.9 8.0 ..
Korea .. .. 4.4 4.1 4.4 4.2 4.8 5.4 5.3 5.5 5.6

ti
ar
Luxembourg .. .. 3.1 4.3 5.2 5.2 5.4 5.6 5.8 6.4 6.8 7.7 8.0
Mexico .. .. .. .. .. .. 4.8 5.6 5.6 6.0 6.2 6.3 6.5

M
Netherlands .. .. 6.6 6.9 7.2 7.1 7.7 8.1 7.9 8.3 8.9 9.1 9.2
New Zealand .. .. 5.1 6.5 5.9 5.1 6.9 7.2 7.7 7.8 8.2 8.0 8.4

er
Norway 2.9 3.4 4.4 5.9 7.0 6.6 7.7 7.9 8.5 8.9 9.9 10.1 9.7

vi
Poland .. .. .. .. .. ..Xa 4.9 5.6 5.7 6.0 6.6 6.5 6.5
Portugal 2.6 5.4 5.6 6.0 6.2 8.2 9.4 9.3 9.5 9.8 10.0
Slovak Republic .. .. .. .. .. .. .. 5.8 5.5 5.5 5.6 5.9 ..
Spain 1.5 2.5 3.5 4.6 5.3 5.4 6.5 7.4 7.2 7.2 7.3 7.9 8.1

c

Sweden .. .. 6.8 7.6 9.0 8.6 8.3 8.1 8.4 8.7 9.1 9.3 9.1
Switzerland 4.9 4.6 5.5 7.0 7.4 7.8 8.3 9.7 10.4 10.9 11.1 11.5 11.6
s

Turkey .. .. .. 3.0 3.3 2.2 3.6 3.4 6.6 7.5 7.4 7.6 7.7
ic

United Kingdom 3.9 4.1 4.5 5.5 5.6 5.9 6.0 7.0 7.3 7.5 7.7 7.9 8.3
om

United States 5.1 5.6 7.0 7.9 8.8 10.1 11.9 13.3 13.3 14.0 14.7 15.2 15.3
on

Source: OECD HEALTH DATA 2006

Ec

18.0!
Australia!
lth

Austria!
Belgium!
ea

16.0! Canada!
Czech Republic!
H

Denmark!
14.0! Finland!
of

France!
Germany!
Greece!
es

12.0!
Hungary!
Iceland!
pl

Ireland!
10.0!
ci

Italy!
Japan!
in

Korea!
Pr

8.0! Luxembourg!
Mexico!
Netherlands!
New Zealand!
6.0!
Norway!
Poland!
Portugal!
4.0! Slovak Republic!
Spain!
Sweden!
2.0! Switzerland!
Turkey!
United Kingdom!
United States!
0.0!
1960! 1965! 1970! 1975! 1980! 1985! 1990! 1995! 2000! 2001! 2002! 2003! 2004!

2-d
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G

2-e
ira
lt
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G

2-f
ira
lt
HEALTH: SPENDING AND RESOURCES
Health spending and financing
Public Pharmaceutical
Averag
Total expenditure as expenditure as
e Health expenditure
expenditure as % of total % of total
growth Per capita USD PPP
% of GDP expenditure on expenditure on
rate
health health
1998-

lt
2003 1993 2003 1993 2003 1993 2003 1993

ira
2003
Australia 1

G
a a d a e
9.3 8.2 68 66 4.1 2 699 1 542 14 10

z-
Austria 7.6 a
7.8 70 a
74 1.8 d
2 280 a
1 669 16 a
11 h

ne
Belgium 9.6 8.1 .. .. 4.2 2 827 1 601 17 f
17

ti
Canada 9.9 b
9.9 70 b
73 4.2 3 003 b
2 014 17 13

ar
Czech Republic 7.5 6.7 90 95 5.4 1 298 760 22 19

M
Denmark 9 8.8 83 83 2.8 2 763 1 763 9.8 8.5

er
Finland 7.4 8.3 | 77 76 | 4.1 2 118 1 430 | 16 12
France
vi
b b
Xa b
10 9.4 76 77 3.5 2 903 1 878 21 18
Germany 11 9.9 78 80 1.8 2 996 1 988 15 13

c

Greece 9.9 8.8 51 b

55 4.9 2 011 1 077 16 17
s

Hungary
ic

7.8 a
7.7 70 a
87 6 | 1 115 a
638 28 a
28
om

Iceland 11 b
8.4 84 b
83 5.9 3 115 b
1 745 15 12
Ireland a a
11 | a a
on

7.3 7 75 73 2 386 1 039 11 11

Italy 8.4 8 75 76 3.1 2 258 1 529 22 20
Ec

Japan 7.9 a,b

6.5 82 a,b
79 3 | 2 139 a,b
1 365 18 a
22
lth

Korea 5.6 4.3 49 36 10 1 074 453 29 31

Luxembourg
ea

6.1 a
6.2 85 a
93 5.3 | 3 190 a
1 891 12 a
12 i

Mexico 6.2 5.8 46 43 4 583 397 21 ..

H

Netherlands 9.8 8.6 62 74 4.6 2 976 1 701 11 11

of

New Zealand 8.1 7.2 79 77 3.4 1 886 1 115 14

f
15
es

Norway 10 b
8 84 b
85 5.3 3 807 b
1 695 9.4 a
9.6
pl

Poland 6 a
5.9 72 a
74 3 | 677 a
378 .. ..
ci

Portugal 9.6 7.3 70 63 3.7 1 797 881 23 g

26
in

Slovak Republic 5.9 .. 88 .. 4.1 777 .. 39 ..

Pr

Spain 7.7 7.5 71 77 2.6 1 835 1 089 22 19 h

Sweden 9.2 a
8.6 | 85 a
87 | 5.4 | 2 594 a
1 644 | 13 a
11
Switzerland 12
b
9.4 59
b
54 2.8 3 781
b
2 401 11 9.7
Turkey 6.6 c
3.7 63 c
66 .. 452 c
200 25 c
32 i

United Kingdom 7.7

a
6.9 83
a
85 5.7 | 2 231
a
1 232 16
f
15
United States 2 15 13 44 43 4.6 5 635 3 357 13 8.6
Source:: OECD Health Data 2005

2-g
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G

2-h
ira
lt
Public expenditure on health, % total expenditure on health

1960 1965 1970 1975 1980 1985 1990 1995 2000 2001 2002 2003 2004
Australia 50.4 50.9 57.2 73.1 63.0 71.4 67.1 66.7 68.9 67.8 68.1 67.5 ..
Austria 69.4 70.3 63.0 69.6 68.8 76.1 73.5 69.3 69.9 69.5 70.5 70.3 70.7
Belgium .. .. .. .. .. .. .. 78.5 75.8 76.4 75.0 71.1 ..
Canada 42.6 51.9 69.9 76.2 75.6 75.5 74.5 71.4 70.3 69.9 69.6 70.1 69.8
Czech Republic .. .. 96.6 96.9 96.8 92.2 97.4 90.9 90.5 89.9 89.7 89.8 89.2

lt
Denmark .. .. 83.7 85.4 87.8 85.6 82.7 82.5 82.4 82.7 82.9 .. ..

ira
Finland 54.1 66.0 73.8 78.6 79.0 78.6 80.9 75.6 75.1 75.9 76.1 76.2 76.6
France 62.4 71.2 75.5 78.0 80.1 78.5 76.6 76.3 75.8 75.9 78.1 78.3 78.4

G
Germany .. .. 72.8 79.0 78.7 77.4 76.2 80.5 78.6 78.4 78.6 78.2 ..

z-
Greece .. .. 42.6 55.6 59.9 53.7 52.0 52.6 55.5 54.1 53.6 52.8
Hungary .. .. .. .. .. .. 89.1 84.0 70.7 69.0 70.2 72.4 72.5

ne
Iceland 66.7 63.1 66.2 87.1 88.2 87.0 86.6 83.9 82.6 82.7 83.2 83.5 83.4

ti
Ireland 76.0 76.2 81.7 79.0 81.6 75.7 71.9 71.6 73.3 75.6 75.2 78.0 79.5

ar
Italy .. .. .. .. .. 77.6 79.1 71.9 73.5 75.8 75.4 75.1 76.4
Japan 60.4 61.4 69.8 72.0 71.3 70.7 77.6 83.0 81.3 81.7 81.5 81.5 ..

M
Korea .. .. .. .. 33.4 35.8 38.5 35.3 46.2 51.9 50.6 50.7 51.4
Luxembourg .. .. 88.9 91.8 92.8 89.2 93.1 92.4 89.3 87.9 90.3 90.6 90.4

er
Mexico .. .. .. .. .. .. 40.4 42.1 46.6 44.9 43.9 44.1 46.4

vi
Netherlands .. .. 60.2 67.9 69.4 70.8 67.1 71.0 63.1 62.8 62.5 63.0 62.3
New Zealand .. .. 80.3 73.7 88.0 87.0 82.4 77.2 78.0 76.4 77.9 78.3 77.4
Xa
Norway 77.8 80.9 91.6 96.2 85.1 85.8 82.8 84.2 82.5 83.6 83.5 83.7 83.5
Poland .. .. .. .. .. .. 91.7 72.9 70.0 71.9 71.2 69.9 68.6

c

Portugal .. .. 59.0 58.9 64.3 54.6 65.5 62.6 72.5 71.5 72.2 72.6 71.9
Slovak Republic .. .. .. .. .. .. .. 91.7 89.4 89.3 89.1 88.3 ..
s

Spain 58.7 50.8 65.4 77.4 79.9 81.1 78.7 72.2 71.6 71.2 71.3 70.4 70.9
ic

Sweden .. .. 86.0 90.2 92.5 90.4 89.9 86.6 84.9 84.9 85.1 85.4 84.9
om

Switzerland .. .. .. .. .. 50.3 52.4 53.8 55.6 57.1 57.9 58.5 58.4

Turkey .. .. .. 50.0 29.4 50.6 61.0 70.3 62.9 68.2 70.4 71.6 72.1
United Kingdom 85.2 85.8 87.0 91.1 89.4 85.8 83.6 83.9 80.9 83.0 83.4 85.4 85.5
on

United States 23.4 22.7 36.5 41.1 41.3 39.8 39.7 45.3 44.0 44.8 44.8 44.6 44.7
Ec

Source: OECD HEALTH DATA 2006

lth

120.0! Australia!
Austria!
ea

Belgium!
Canada!
H

Czech Republic!
100.0! Denmark!
of

Finland!
France!
Germany!
es

80.0! Greece!
Hungary!
pl

Iceland!
ci

Ireland!
Italy!
in

60.0! Japan!
Korea!
Pr

Luxembourg!
Mexico!
40.0! Netherlands!
New Zealand!
Norway!
Poland!
Portugal!
20.0! Slovak Republic!
Spain!
Sweden!
Switzerland!
Turkey!
0.0! United Kingdom!
1960! 1965! 1970! 1975! 1980! 1985! 1990! 1995! 2000! 2001! 2002! 2003! 2004! United States!

2-i
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G

2-j
ira
lt
Total expenditure on pharmaceuticals (% TEH)
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Australia 9.5 9.9 10.4 11.0 11.2 11.5 11.7 12.0 12.6 13.5 14.0 14.2
Austria 9.2 9.3 11.1 12.1 12.7 12.6 12.3 12.8 13.1 13.0
Belgium 15.6 16.3 17.4 17.5 16.8 16.2 16.5 11.3
Canada 11.8 12.4 13.0 13.1 13.8 14.0 14.8 15.2 15.5 15.9 16.2 16.7 17.0 17.7
Czech Republic 18.4 21.1 19.4 24.7 25.1 25.0 24.9 22.9 23.0 22.4 21.5 22.0
Denmark 8.0 7.9 8.5 8.8 9.1 8.9 9.0 9.0 8.7 8.8 9.2 9.8 10.0 9.4
Finland 9.9 10.8 12.3 13.4 14.1 14.4 14.8 14.6 15.0 15.5 15.8 16.0 16.0 16.3

lt
France 17.2 17.1 17.5 17.4 17.6 17.6 18.0 18.6 19.5 20.3 20.9 18.7 18.8 18.9

ira
Germany 14.7 13.2 12.9 12.7 12.8 12.9 13.4 13.5 13.6 14.2 14.5 14.6
Greece 16.3 17.0 16.6 16.1 15.7 16.1 16.2 13.9 14.4 15.0 15.1 16.2 17.1 17.4

G
Hungary 27.6 26.5 28.4 28.0 25.0 26.0 25.9 28.5 27.6

z-
Iceland 12.3 13.0 12.4 13.1 13.4 14.0 15.1 14.8 14.3 15.1 14.4 14.7 14.5 14.8

ne
Ireland 11.6 11.1 10.7 10.6 10.4 10.5 10.2 10.4 10.5 10.6 10.3 11.0 11.8 12.4
Italy 20.8 20.8 20.2 20.3 21.1 21.3 21.5 22.0 22.6 22.4 22.6 22.5 22.1 21.4

ti
Japan 22.9 22.0 22.3 21.1 22.3 21.6 20.6 18.9 18.4 18.7 18.8 18.4 18.9

ar
Korea 35.0 33.3 32.3 31.3 31.4 30.2 27.7 25.8 25.1 27.8 27.6 27.9 27.6 27.4

M
Luxembourg 15.0 12.2 12.0 11.5 12.6 12.3 11.9 11.0 11.5 10.3 9.4 8.5
Mexico 18.6 19.4 19.6 21.2 21.5 20.9

er
Netherlands 9.6 10.5 11.0 10.9 11.0 11.0 11.0 11.2 11.4 11.7 11.7 11.5

vi
New Zealand 14.1 14.2 14.9 15.8 14.8 14.5 14.4 Xa
Norway 7.3 7.5 9.6 8.8 9.0 9.1 9.1 8.9 8.9 9.5 9.3 9.4 9.2 9.5
Poland 28.4 30.3 29.6

Portugal 24.3 24.7 25.6 25.2 23.6 23.8 23.8 23.4 22.4 23.0 23.3 22.6 23.2
c

Slovak Republic 34.0 34.0 34.0 37.3 38.5

s

Spain 19.2 19.8 20.8 21.0 21.5 21.3 21.1 21.8 22.8 22.8
ic

Sweden 8.7 9.7 10.7 11.8 12.3 13.6 12.4 13.6 13.9 13.8 13.2 13.0 12.6 12.3
om

Switzerland 9.8 9.4 9.7 9.8 10.0 10.0 10.3 10.2 10.5 10.7 10.6 10.3 10.5 10.4
Turkey 31.6 24.3 24.8
on

United Kingdom 13.8 14.2 14.8 15.1 15.3 15.6 15.8

United States 9.0 8.7 8.5 8.5 8.9 9.3 9.7 10.3 11.1 11.7 12.0 12.3 12.4 12.3
Ec

Source: OECD HEALTH DATA 2006

lth

45.0!
Australia!
ea

Austria!
40.0! Belgium!
H

Canada!
Czech Republic!
Denmark!
of

35.0! Finland!
France!
es

Germany!
30.0! Greece!
pl

Hungary!
Iceland!
ci

25.0! Ireland!
Italy!
in

Japan!
20.0! Korea!
Pr

Luxembourg!
Mexico!
15.0! Netherlands!
New Zealand!
Norway!
10.0! Poland!
Portugal!
Slovak Republic!
5.0! Spain!
Sweden!
Switzerland!
Turkey!
0.0! United Kingdom!
1991! 1992! 1993! 1994! 1995! 1996! 1997! 1998! 1999! 2000! 2001! 2002! 2003! 2004! United States!

2-k
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G

2-l
ira
lt
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G
ira

2-m
lt
 lt
ira
G
z-
ne
ti
ar
M
er
Reading:
vi
Xa

c

Doyle, J., 2007, Returns to Local-Area Health Care

s
ic
om

Spending: Using health stocks to patients far from

on

home, NBER WP 13301

Ec
lth
ea

[http://www.nber.org/papers/w13301]
H
of
es
pl
ci
in
Pr

2-n
 National Bureau of Economic Research
HTTP://WWW.NBER.ORG/

Higher Health Spending Saves Lives

"A typical comparison of a high-spending area and a low-spending one means a 50 percent difference in health care spending
intensity�This disparity is associated with a 1.6 percentage-point lower mortality rate among heart emergency patients. Based
on that estimate, the additional cost of a statistical life-year-saved is on the order of $50,000."

lt
ira
Health care spending is a major concern in the United States, amounting to over $2 trillion per year or 16 percent of GDP.
These figures are expected to increase with the aging of the population and are likely to strain government budgets and

G
private-sector profitability. And, there is controversy over exactly what we are getting for that health care spending.

z-
Among counties or regions within the United States, there are large disparities in spending, yet health outcomes are

ne
remarkably similar. One study of Medicare data found that end-of-life spending levels -- a measure of treatment intensity
that controls for the health outcome -- are 60 percent higher in high spending areas of the United States than in low

ti
spending areas. Yet there is no difference across regions in five-year mortality rates after such health events as heart attacks

ar
or hip fractures.

M
One difficulty that arises when comparing regions is that populations in worse health may receive greater levels of treatment.
For example, at the individual level higher spending is strongly associated with higher mortality rates, because more is spent

er
on sicker patients. At the regional level, long-term investments in capital and labor also may reflect the underlying health of
the population.

vi
Xa
In Returns to Local-Area Health Care Spending: Using Health Shocks to Patients Far From Home (NBER Working
Paper No. 13301), author Joseph Doyle compares outcomes of patients who are exposed to different health care systems

that were not designed for them: patients who are far from home when a health emergency strikes. These visitors vacation
c

in areas that provide different levels of health care. They may have a health emergency in an area that spends a great deal
s

on patients or in one that tends to spend less. By comparing similar visitors across these locations, Doyle is able to use
ic

differences in health outcomes to shed light on the returns to health care spending, at least in emergency situations.
om

He finds that if the medical emergency occurred in a high-spending area, the patient was significantly more likely to survive.
This result comes from analyzing groups of counties with similar lodging prices that are also popular tourist destinations --
on

areas that are likely to be close substitutes in terms of vacations, and that provide credible variation in health care systems.

In particular, Doyle uses data from hospital discharges in the state of Florida -- one of the most frequently visited states,
Ec

which also gathers a wealth of data on patient characteristics. A typical comparison of a high-spending area and a low-
spending one means a 50 percent difference in health care spending intensity. Doyle finds that this disparity is associated
lth

with a 1.6 percentage-point lower mortality rate among heart emergency patients. Based on that estimate, the additional cost
of a statistical life-year-saved is on the order of $50,000 -- similar to the estimate from health improvements over time, and
ea

well below the typical value of a life-year-saved of $100,000.

H

Doyle's results also confirm earlier findings of little relationship between spending and mortality among the populations the
health care systems are designed to serve. Instead, those who have a serious health emergency far from home are exposed
of

to different health care systems, but they are unlikely to affect the resources available in the systems.
es

Doyle points out that visitors choose their destinations, and if relatively healthy individuals were to choose high-spending
areas, then his main results would reflect these differences. However, his estimates are robust across different types of
pl

patients, including those with various income levels, and within groups of destinations that can be characterized as close
ci

substitutes. The returns to spending are lower in places where the visitors were more likely to select the destination with the
health care system in mind -- this suggests that Doyle's main results may understate the benefits of health care spending.
in
Pr

-- Les Picker

The Digest is not copyrighted and may be reproduced freely with appropriate attribution of source.

2-o
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G

2-p
ira
lt
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G

2-q
ira
lt
Why is the health care market different?

⊕ difference between the health status a consumer

wants and the level of medical care that is able to
buy;

lt
ira
difficulty to measure the product and its costs;

G
z-
⊗ patient does not decide treatment. Physician acts

ne
on his behalf (consumer’s agent)

ti
ar
⊗ patient does not pay treatment → health insur-

M
er
ance
vi
 traditional mechanisms to limit market power of
Xa

c

suppliers do not work because

s
ic
om

? possibility of insurance (moral hazard and ad-

on

verse selection)
Ec
lth

? entry barriers (professional associations, ex-

ea

ams to access specialized practice, ...).

H
of
es

Also, external factors contribute:

pl
ci
in

† population aging,
Pr

‡ technological development.

Consequence: MARKET INTERVENTION. [−→no

guarantee of proper behavior!!]
2-r
 lt
ira
Differential characteristics

G
z-
ne
- presence of uncertainty - (sec. 8)

ti
ar
M
- relevance of insurance - (sec 9)

er
- presence of asymmetric information - (sec 9)
vi
Xa
- role of non-profit institutions - (sec 6)

c

- extent of regulation in the market - (sec 4)

s
ic
om

- existence of “need”
on

- public provision and financing of health care ser-

Ec

vices
lth
ea

- presence of externalities and ”merit goods” - (sec

H

4)
of
es
pl
ci

- (moral issue) universal access to the health care

in
Pr

system

2-s
The elements of health economics
F. MICROECONOMIC APPRAISAL
Cost-eFfectivenes, Cost-benefit, Cost-utility
Analysis of alternative ways of delivering care
(mode, place, timing or amount) at all phases
(detection, diagnsis, treatment, etc.)
E. MARKET ANALYSIS
Money prices, time prices; waiting lists and non-
price rationing systems as equilibrating
mechanisms and their differential effects in markets
for physician and hospital services.

lt
ira
G
z-
ti ne
B. WHAT INFLUENCES HEALTH A. WHAT IS HEALTH? WHAT IS ITS

ar
(OTHER THAN HEALTH CARE)? VALUE?
G en etics , o ccu p atio n al h azar d s ; Perceived attributes of health; health status

M
consumption patterns; education; income; indices; value of life; utility scaling oh

er
capital (human and physical), family health.
background, etc.

vi
Xa

c
s

C. DEMAND FOR HEALTH CARE D. SUPPLY OF HEALTH CARE

ic

Influence of A and B on health care Production costs, alternative production

om

seeking behavior; barriers to care techniques, input substitution; markets for

seeking; agency relationship; need; inputs; remuneration methods and incentives;
on

altruism; insurance; demand for care forprofit and nonprofit organizations.

and its effects.
Ec
lth
ea
H

G. PLANNING, BUDGETING, H. EVALUATION AT THE WHOLE SYSTEM

of

REGULATION, AND MONITORING LEVEL

MECHANISMS Equity and allocative eficiency criteria brought to
es

Evaluation of effectiveness of instruments bear on E and F; inter-regional and international

pl

available for optimizing the system; inter- comparisons of performance; financing methods.
ci

play of budgeting, manpower allocations,

regulation, and their incentive structures.
in
Pr

Economic analysis: Starting point: A →

Demand side Supply side Equilibrium
C (with B) D E (from B, C)
Economic evaluation: microlevel (F ), macrolevel (G)
Policy analysis: H
2-t
The organization of the health care market

lt
ira
G
z-
ne
ti
ar
M
er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

The agents in a health care system

3
Structure of the health care system

lt
ira
G
z-
ne
ti
ar
M
er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

Private provision with and without insurance

3-a
The reimbursement model

Public version (France) and private version (UK, The

lt
ira
Netherlands).

G
z-
Separation between providers and 3PP.

ne
ti
Patient advances payment and is reimbursed (par-

ar
M
tially or totally) by 3PP .

er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

The reimbursement model

3-b
The contract model

lt
ira
Link between providers and 3PP.

G
Public version (primary care UK, Ireland, Germany,

z-
ne
Netherlands) and private version.

ti
ar
Patients choose providers among in-plan providers.

M
er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

The contract model

3-c
The integrated model

lt
ira
Integration of providers and 3PP.

G
3PP contract physicians and own hospitals.

z-
ne
Public version (Spain, Portugal, Ireland, hosp. care

ti
ar
UK), and private version (US-HMOs).

M
er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

The integrated model

3-d
The actors of the health care market

lt
ira
G
z-
ne
ti
ar
M
er
vi
Xa

c
s
ic
om
on

- Market of insurers (3PP): Private vs Public

Ec
lth
ea

- Market of hospital services: Entry barriers, con-

H
of

centration, Ecs. of scale, Regulation

es
pl
ci

- Market of physician services: Supplier-induced de-

in
Pr

mand, Competitiveness, Costs and quality

- Also Regulation: reimbursement systems to providers

4
 lt
ira
G
z-
ne
Flows of resources in industrial economics

ti
ar
M
er
vi
Xa
Prices / production Gross
Revenues

c

Markets
s
ic

Demand Supply Costs

Profits
om
on
Ec

Acquisitions Advertising R&D Capital Retained Dividends

lth

Marketing investment earnings

ea
H
of

Funds Stock
External market
available
es

funds valuation
pl
ci
in
Pr

4-a
 lt
Financial flows in industrial economics

ira
G
z-
ti ne
Gross

ar
Prices / production
Revenues

M
Markets

er
vi
Demand Supply Costs
Profits
Xa

c
s
ic

Acquisitions Advertising R&D Capital Retained Dividends

om

Marketing investment earnings

on
Ec

Funds Stock
External
lth

available market
funds valuation
ea
H
of
es

- Generation of resources
pl
ci
in
Pr

- Distribution of profits

4-b
 lt
ira
G
Expenditure flows in industrial economics

z-
ti ne
ar
M
Gross

er
Prices / production
Revenues
Markets
vi
Xa

Costs
c

Demand Supply Profits

s
ic
om

Acquisitions Advertising R&D Capital

on

Retained Dividends
Marketing investment earnings
Ec
lth
ea

Funds Stock
External market
H

available funds valuation

of
es
pl
ci

- Use of available flows

in
Pr

4-c
 lt
ira
G
Market flows in industrial economics

z-
ti ne
ar
M
er
Prices / production Gross

vi
Xa Revenues
Markets

c

Demand Supply Costs

Profits
s
ic
om
on

Acquisitions Advertising R&D Capital Retained Dividends

Ec

Marketing investment earnings

lth
ea

Funds Stock
External
H

market
available funds
of

valuation
es
pl
ci

- Generation of profits
in
Pr

4-d
 lt
ira
2. The agents of the economy

G
z-
ne
ti
Population/ Patients (Demand)

ar
M
er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

5
 Pr
in
ci
pl
es
of
H
ea
lth
Ec 3D consumption set
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G

5-a
ira
lt
 lt
ira
Individual vs. aggregate demand

G
z-
ne
ti
Individual demand → solution of

ar
M
er
max U (x, y) s.t. M = Pxx + Py y
x,y
vi
Xa

c

x∗(Px, Py , M )
s
ic

y ∗(Px, Py , M )
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

5-b
 U

lt
ira
G
U (x1 , y1 )

z-
ne
y1

ti
ar
M
er
vi
Xa
x1

c
s
ic
om

x
on
Ec

y
lth
ea

M M Px
y= − x
Py Py
H

Py
of
es
pl
ci
in

y∗
Pr

u3

!u
u2

u1

x∗ M x
Px

5-c
Consider 2 individuals x1(Px, Py , M1) and
x2(Px, Py , M2).

lt
ira
G
z-
ne
The aggregate (market) demand for good x is the

ti
ar
horizontal sum of individual demands.

M
er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

5-d
Effects on (aggregate) demand

Changes along the demand curve [(Py , M ) given]

lt
ira
G
z-
- ↑ Px, x ↓: some consumers buy less and some

ne
ti
others leave the market.

ar
M
er
vi
- ↓ Px, x ↑: some consumers buy more and some
Xa

others enter the market.

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in

Shifting the demand curve [(Px, Py ) given]

Pr

- ↑ M −→ increase demand x and y: demand curve

moves outwards.

5-e
Crossed effects [(Px, M ) given]

Impact of ↑ Py (M constant) on x, three possibili-

ties:

lt
ira
G
(i) x and y independent, e.g. (x,y)= (coffee, gaso-

z-
ne
line):

ti
↑ Py →↓ y → demand of x unaffected

ar
M
er
vi
(ii) x and y substitutes: satisfy similar needs, e.g.
Xa

(x,y) = (butter, margarine):

c
s

↑ Py →↓ demand of y →↑ demand of x.
ic
om
on
Ec

(iii) x and y complements: joint consumption, e.g.

lth

(x,y) = (coffe, sugar):

ea

↑ Py →↓ demand of y →↓ demand of x.
H
of

Px Px
es

M given M given
pl

!Py !Py
ci
in
Pr

Px Px

x1 x2 x x2 x1 x

(x, y) substitutes (x, y) complements

5-f
Elasticity

lt
ira
How to measure the impact of ∆Px on x?

G
z-
ne
Method 1: Direct and simple

ti
ar
M
∆x

er
∆Px
vi
Xa

Problem: dependent on units

c
s
ic
om

EURO US $
on

Px x Px x
Ec

6 10 8 10
lth
ea

12 5 16 5
H
of
es
pl

∆x  −5
ci


= −0.83
in

=
Pr

∆Px EU R

6
∆x  −5
= = −0.625
∆Px $ 8

5-g
Method 2: Index invariant to units −→
Elasticity

lt
ira
Own-price elasticity

G
z-
 ∆x 
 %∆x   ∆xPx 
   

ne
 x 
|εx| =   ∆Px  = 
=

ti
  
%∆Px ∆P x

ar
 
Px x

M
er
|εx| > 1 elastic (overreaction)
vi
Xa

c

|εx| < 1 inelastic (underreaction)

s
ic
om
on

Example: |εx| = 1
2
Ec
lth
ea

Cross-price elasticity
H
of

∆x
%∆x x = ∆xPy ≶ 0 compl v. subs
es

εxy = = ∆P
pl

%∆Py y ∆Py x
ci

Py
in
Pr

Income elasticity
∆x
%∆x x ∆xM
ηx = = ∆M = >0
%∆M M
∆M x

5-h
Ilustration

lt
ira
Derivation of the demand function

G
z-
ne
- Consider a two-good economy: a composite con-

ti
ar
M
sumption good (y) [“food”] and health care (x).

er
- (Representative) individual’s utility function:
vi
Xa
U (x, y) = xαy β , α, β > 0

c
s
ic

- Individual’s income m.
om
on

- Individual’s budget constraint:

Ec
lth

m ≥ xPx + yPy
ea
H

where Px y Py denote prices of x and y respectively.

of
es
pl

- Individual’s problem:
ci
in

Select a bundle (x, y) to maximize utility given (Px, Py ; m):

Pr

max xαy β s.t. m ≥ xPx + yPy

x,y

5-i
Solution:
max L(x, y) = xαy β + λ(m − xPx − yPy )
x,y
First order conditions,

lt
ira
∂L
= αxα−1y β − λPx = 0 (1)

G
z-
∂x

ne
∂L
= βy β−1xα − λPy = 0

ti
(2)

ar
∂y

M
∂L

er
= m − xPx + yPy = 0 (3)
∂λ vi
Xa

From (1) and (2),

c
s
ic

αy Px
om

=
βx Py
on
Ec

That is,
lth

βx Px
ea

y= (4)
H

α Py
of
es

Substituting (4) in (3) yields

pl

αm
ci

x(Px, m) = (5)
in

Px(α + β)
Pr

Substituting (5) in (4) yields

βm
y(Py , m) = (6)
Py (α + β)
Note: indep goods! ∂x/∂Py = 0 = ∂y/∂Px

5-j
Example Society with two consumers a and b and
two goods x and y.

lt
ira
1 2

G
x y3
Ua(xa, ya) = a a3

z-
ne
2 1
x y3
3

ti
Ub(xb, yb) = b b

ar
M
Individual demands:

er
mvi
Xa
xa(Px, m) =

c

3Px
s

2m
ic

ya(Py , m) =
om

3Py
on

2m
Ec

xb(Px, m) =
lth

3Px
m
ea

yb(Py , m) =
H

3Py
of
es

Market demands:
pl
ci

m
in

x(Px, m) =
Pr

Px
m
y(Py , m) =
Py

5-k
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G

5-l
ira
lt
Elasticity

lt
ira
G
♦ own-price elasticity

z-
ne
ti
∂xa Px 1

ar
εx a = =−

M
∂Px xa 3

er
∂xb Px 2
εxb = =−vi
Xa
∂Px xa 3

c

∂x Px
s

εx = = −1
ic

∂Px xa
om
on
Ec

♦ income elasticity
lth
ea

∂xa m
H

ηxa = =1
∂m xa
of
es

∂xb m
ηxb = =1
pl
ci

∂m xa
in

∂x m
Pr

ηx = =1
∂m xa

5-m
Example: Manning, W. and C. Phelps, 1979, The
demand for dental care, Bell Journal of Economics,
10: 503-525.

lt
ira
G
z-
ne
ti
ar
M
er
vi
Xa

c

Price and Income elasticities of demand for primary dental services.

s
ic
om

Comments:
on
Ec

1. Price elasticity of demand for children much greater

lth

than adults.
ea
H
of

2. Children demand for cleanings, fillings and ex-

es

tractions quite sensitive to price .

pl
ci
in
Pr

3. Some income elasticities are substantial in clean-

ings and examinations.

4. Negative income elasticity for extractions in adults:

“poor people’s dentistry”.
5-n
Readings:

lt
ira
G
z-
Manning, W. and C. Phelps, 1979, The demand for

ne
ti
dental care, Bell Journal of Economics, 10: 503-

ar
M
525.

er
vi
Xa
[http://ideas.repec.org/a/rje/bellje/v10y1979iautumnp503-525.html]

c
s
ic
om

AIHW, 2003, Demand for dental care, AIHW Dental

on

Statistics and Research Unit Research Report No.

Ec

8.
lth
ea
H

[http://arcpoh.adelaide.edu.au/publications/report/research/]
of
es
pl

Note: Usually dental care is not covered by health

ci
in

insurance. Demand for dental care is thus sensitive

Pr

to price variations.

5-o
Producers (Suppliers).

lt
PRODUCERS What and how much to produce

ira
G
z-
Production function

ne
Possibilities Technology Technological costs

ti
ar
- Total

Economic costs
M
- Average
- Marginal

er
vi
Xa

Supply Feasible production set

c

Opportunity costs.
Efficiency
s
ic

PPF
om

input 1
on

Production possibility frontier

Ec

C(x) = F + V (x)
lth

C(x)
ea

Opportunity AC(x) =
x
H

set
∂C(x)
of

MC(x) =
input 2 ∂x
es
pl
ci

output
in

€
Pr

Production function

input

6
Production function

♠ relation between output and inputs: output = f(inputs).

lt
ira
→ engineering approach to production activity.

G
bread=f(flower, water, salt, labor, ...)

z-
ne
surgery=f(surgery room, blood, anesthesia, nurse,

ti
ar
surgeon, ...)

M
er
e.g. q = f (K, L)
vi
Xa

c

♠ Def.: represents the maximum amount of output

s
ic
om

that can be obtained from a given combination of

on

inputs. (conveys efficiency)

Ec
lth
ea

♠ Graphical representation (1 output, 2 inputs):

H
of
es

(a) isoquant map → degree of substitutability of in-

pl
ci

puts.
in
Pr

(b) 3D

(c) Production possibility frontier (multiproduct)

6-a
(a) isoquant map

input 1

lt
ira
G
q3 > q2 > q1

z-
α

ne
A

ti
ar
M
q3

er
vi
q2
Xa
γ q1

c

β input 2
α
tg γ = − (degree of input substitutability)
s

β
ic
om
on

(b) 3D representation
Ec
lth

output
ea
H
of

input 1
es
pl
ci
in
Pr

input 2

6-b
Consider a hospital with 10 surgeons and 2 activi-
ties.
If all perform knee surgery → 50 interventions/week;

lt
If all perform hip replacements → 50 interventions/week.

ira
G
z-
ne
hip

ti
ar
50 B
45

M
er
vi
Xa
25
A

c

16 C
s
ic
om
on

knee
16 25 45 50
Ec
lth
ea

Points A, B, C ∈ feasible production set.

H
of

Represent production of hospital (supply).

es

Points B, C ∈ FPP.
pl
ci
in
Pr

Production possibility frontier:

Set of all the maximum combinations of operations
the hospital can achieve given the quantity and pro-
ductivity of resources available.

6-c
Efficiency.

lt
ira
An allocation of resources is efficient if it is impos-

G
z-
sible to change that allocation to make a consumer

ne
ti
better off (perform one additional intervention) with-

ar
M
out making anybody else worse off (reducing num-

er
ber of operations). vi
Xa

c
s
ic

Efficiency refers to allocations of resources yielding

om

the maximum possible output, i.e. allocations on

on
Ec

PPF.
lth
ea

Hence, allocation A is not efficient, while allocations

H
of

B, C are efficient.
es
pl
ci
in

From a social point of view, there is interest in mov-

Pr

ing from A to B (or C). The hospital is able to in-

crease its output with the same inputs.

6-d
Efficacy.

Potential benefit of a technology. Probability that an

lt
ira
individual benefits from the application of a (health)

G
technology to solve a particular (health) problem,

z-
ne
under ideal conditions of application.

ti
ar
M
er
Effectiveness.
vi
Xa

c

Probability that an individual benefits from the appli-

s
ic
om

cation of a (health) technology to solve a particular

on

(health) problem, under real conditions of applica-

Ec

tion.
lth
ea
H

Examples:
of
es
pl
ci

Highly effective treatments: vaccinations, heart surgery,

in
Pr

diabetes, influenza, renal insufficiency, ...

Clinical interventions of known efficacy explain 5 of

the years won in life expectancy at birth.

6-e
 lt
ira
Efficacy vs Effectiveness

G
z-
ne
In general, efficacy or ideal use or perfect use is

ti
ar
the ability to produce a specifically desired effect.

M
er
For example, an efficacious vaccine has the ability
vi
Xa
to prevent or cure a specific illness. In medicine a

c

distinction is often drawn between efficacy and ef-

s
ic
om

fectiveness or typical use. Whereas efficacy may

on

be shown in clinical trials, effectiveness is demon-

Ec

strated in practice.
lth
ea

The distinction between efficacy and effectiveness is important

H

because doctors and patients often do not follow best practice

of

in using a treatment. For instance, a patient using oral con-

es

traceptive pills to prevent pregnancy may sometimes forget to

pl

take a pill at the prescribed time; thus, while the perfect-use

ci
in

failure rate for this form of conception in the first year of use is
Pr

just 0.3%, the typical-use failure rate is 8%.

6-f
Illustration

F Clinical essay: efficacy of drug 1=75%.

F ∃ drug 2, same price and efficacy = 70%

lt
ira
G
z-
more effective to select drug 1 and reject drug 2?

ne
ti
ar
M
YES, with this information.

er
vi
Xa
Additional INFO

c
s
ic
om

 both drugs are correctly prescribed to 75% of pa-

on

tients
Ec

 drug 1: 50% of patients follow treatment correctly

lth
ea

(shots)
H

 drug 2: 70% of patients follow treatment correctly

of
es

(pills)
pl
ci
in
Pr

Effectiveness of drugs:

E1 = 0.75 × 0.75 × 0.5 = 0.28125

E2 = 0.7 × 0.75 × 0.7 = 0.3675
Conclusion: select drug 2.
6-g
Cost function

Cost function shows relationship between output and

cost. → economic approach to production activity.

lt
ira
G
Def.: minimum possible cost of production of a given

z-
ne
volume of output (q̄). (conveys efficiency)

ti
ar
CT (q̄) = minK,L rK + wL s.t.q = f (K, L)

M
er
→ K(q̄), L(q̄)
vi
Xa

c

Example: Let,
s
ic

q represent physician office visits,

om

L represent labor input (with price w = 1 e),

on

K represent capital input (with price r = 1.2 e).

Ec
lth
ea

We are assuming competitive markets!

H
of
es

Short run vs. long run: fixed costs.

pl
ci

Total cost: T C(q̄) = rK(q̄) + wL(q̄) =

in
Pr

1.2K(q̄) + L(q̄)
Average cost: AC(q̄) = T C(q̄)
q̄
Marginal cost: M C(q̄) = ∂T ∂q
C(q̄)

Representation: Isocost map → K = TrC − w

rL
6-h
 K

T C3 T C3 > T C2 > T C1
r

T C2
r

lt
T C3 w

ira
T C1 K= − L
r r

G
r

z-
ti ne
ar
M
T C1 T C2 T C3

er
T C2 T C3
vi
T C1 Xa L
w w w

c
s

To derive the total cost function, combine isocost

ic
om

map and isoquant map:

on
Ec

- To produce q = 100 (i.e. 100 visits of patients)

lth
ea

given the prices w and r, the physician minimizes

H

cost by contracting 20 units of labor and 25 of capi-

of

tal. This yields a total cost of T C(100) = (1.2)25+

es
pl

20 = 50 e.
ci
in
Pr

- To producte q = 150, → T C(150) = (1.2)40 +

30 = 78 e

- To producte q = 200, → T C(200) = (1.2)50 +

45 = 105 e
6-i
K TC TC(q)
5
10
=
C

C
T

105

lt
78

ira
=
C
T

G
B

z-
C 78
50

ne
B
40 A

ti
50

ar
A q = 200

M
25
50
=

er
q = 150
C
T

vi
Xa
q = 100

20 30 45
c

L 100 150 200 q

s
ic
om

TC AC
on

TC(q) MC
Ec

MC(q)
lth

δ AC(q)
ea

T C(q̃)
H
of
es
pl
ci
in
Pr

γ
q̃ q q

T C(q̃)
AC(q̃) = tg γ = ; M C(q̃) = tg δ
q̃

6-j
Remark 1: decreasing (long run) AC implies a range

lt
ira
of values of q such that M C(q) < AC(q).

G
z-
∂ T C(q)

ne
∂AC(q) M C(q)q − T C(q)
q

ti
= = =

ar
∂q ∂q q 2

M
M C(q) AC(q)

er
− < 0 ⇔ M C(q) < AC(q)
vi
Xa
q q

c
s
ic

Remark 2: let q̂ be such that AC(q̂) is minimum.

om
on

Then, AC(q̂) = M C(q̂).

Ec
lth

If AC(q̂) is minimum means derivative = 0. Thus,

ea
H

T C(q) 
of

∂AC(q) 
 ∂ q  M C(q)q − T C(q) 

es


 =  =  =
pl

∂q q̂ ∂q q̂ q 2 q̂
ci
in

M C(q)  AC(q) 
 
Pr

 −  = 0 ⇔ M C(q̂) = AC(q̂)
q q̂ q q̂

6-k
Economies of scale

lt
ira
G
z-
Economies (diseconomies) of scale characterizes a

ne
ti
production process in which an increase in the level

ar
M
of production causes a decrease (increase) in the

er
long run average cost of each unit.
vi
Xa

c
s

AC
ic
om

AC(q)
on
Ec
Eco

le
no

ca
lth
mie

fs
o
ea
so

ies
om
f sc
H

n
ale

co
of

se
Di
es
pl
ci
in
Pr

optimal size of hospital q

6-l
Economies of scope

Economies of scope may appear in multiproduct firms.

lt
ira
Scope economies refer to changes in average costs

G
z-
induced by changes in the mix of output between

ne
two or more products. In other words, they refer to

ti
ar
M
the potential cost savings from joint production.

er
vi
Xa
Consider a community with two hospitals. One spe-

c

cialized in pediatric care (q1), the other specialized

s
ic
om

in cancer care (q2). May it be worth to merge both

on

activities in a single hospital?

Ec
lth
ea

Scope economies arise if

H
of

T C(q1, q2) < T C(q1) + T C(q2)

es
pl
ci

That is, the joint production of pediatric and can-

in
Pr

cer care allows for savings in the hospital’s manage-

ment structure, administration systems, management
of hospital capacity, nurses, and non-sanitary per-
sonnel, etc.

6-m
Opportunity cost.

lt
ira
G
The concept of opportunity cost is defined as the

z-
ne
benefit given up by not choosing an alternative allo-

ti
ar
cation.

M
er
vi
Xa
Assume a shift from B to C (page 6b). Consequences?

c
s
ic
om

- 29 additional heart surgery interventions

on

- 29 less hip replacements.

Ec
lth
ea

The opportunity cost of moving from B to C is the

H

reduction in hip replacements due to the increase in

of
es

heart operations.
pl
ci
in
Pr

The opportunity cost is an economic concept (not in

accountancy).

6-n
How does society chooses among feasible alloca-
tions? VOTING mechanism.

Criteria to be used:

lt
ira
G
- Efficiency: Select only efficient allocations (rule out

z-
ne
allocation A)

ti
ar
M
er
- Equity. [Normative criterion] Select allocations meet-
vi
ing society’s requirement for justice.
Xa

c

→ people’s values
s
ic

e.g. social justice is behind the set-up of a NHS.

om
on
Ec

FHorizontal and Vertical equity.

lth
ea

 Horizontal equity: equal treatment of equal need.

H
of

 2 individuals with same illness and severity should

es

receive same treatment.

pl
ci
in
Pr

 Vertical equity: unequal treatment of unequal need.

 more treatment for patients with serious condi-
tions than for those with minor affections.
 passing the financing of health care to ability to
pay (progressive income tax).
6-o
Technical progress and its diffusion

Technical progress: Defs.:

lt
(a) produce “old” goods less costly, or produce “new”

ira
G
goods.

z-
ne
ti
(b) Ability to produce at a lower cost given a quality

ar
M
level.

er
vi
Xa
Diffusion: who adopts a new tech, and why.

c
s
ic
om

2 principles:
on

- profit principle: physicians more likely to adopt a

Ec

new surgical technique if it is expected to increase

lth

their revenue stream by enhancing their prestige and/or

ea
H

by improves well-being of patients. [if present value

of

of future profits due to innovation > 0.]

es

- information principle: role of friends, colleagues,

pl
ci

journals, and conferences at informing and encour-

in
Pr

aging the adoption decision.

Trade-off:
- waiting may give rivals a competitive advantage;
- waiting allows for learning from others’ experience.
6-p
(Classic) Pattern of diffusion

- Slow at the beginning;

lt
ira
G
z-
- Then at an increasing rate;

ne
ti
ar
M
- Then at a decreasing rate asymptotically reaching

er
its limit K. vi
Xa

c
s

% adopters
ic
om

K
on
Ec
lth

K
ea

Pt =
1 + e−(a+bt)
H
of
es
pl
ci
in
Pr

time

(a, b) parameters to be estimated.

6-q
Individual vs. aggregate supply

lt
ira
G
z-
Individual supply → solution of

ne
ti
ar
max Π(q) = qPq − C(q)

M
q

er
That is, (w input price vector) vi
Xa

c

q ∗(Pq , w) → market structure?

s
ic
om
on

Pq
Ec

w given
lth
ea
H
of
es
pl

q
ci
in
Pr

NOTE: Pq vs. P (q).

6-r
 lt
ira
G
Consider 2 firms q1(Pq , w) and q2(Pq , w).

z-
ne
The aggregate (market) supply for good q is the hor-

ti
ar
izontal sum of individual supplies.

M
er
vi
Xa

c

Firm 1 Firm 2 Aggregate Supply

Pq
Pq Pq
s
ic

P1
om
on

P2
Ec
lth

P3
ea

P4
H
of

q11 q21 q q12 q22 q q1 q2 q

es
pl

q1 = q11 + q12
ci

q2 = q21 + q22
in
Pr

6-s
Effects on supply

Changes along the supply curve

lt
ira
- ↑ Pq , q ↑: some firms produce more and some

G
others enter the market.

z-
ne
ti
ar
- ↓ Pq , q ↓: some firms produce less and some

M
others leave the market.

er
vi
Xa
Pq Pq Pq

c

M given
s
ic

�w �Tech
om
on
Ec
lth

q q q
ea
H
of

Shifting the supply curve

es
pl
ci

- ↑ w, (Pq constant), same production level is more

in
Pr

expensive −→↓ production: supply moves inwards.

- R&D −→ more efficient technology −→ same pro-

duction level is cheaper −→
↑ production: supply moves outwards
6-t
Illustration

Consider a firm (hospital) with a production function

of health services q(l) = lδ , where l denote work-

lt
ing hours and q health services.

ira
G
z-
ne
The associated cost function C(w, q) = wl(q) where

ti
ar
l(q) = q 1/δ , that is,

M
er
1
C(q, w) = wq δ
vi
Xa

The (competitive) profit function is

c
s
ic

Π(q) = qPq − C(q)

om
on

The problem of the hospital is to determine the level

Ec

of q to maximize profits. Formally,

lth
ea

1
H

max qPq − wq δ (7)

q
of
es

First order condition:

pl
ci

∂Π 1 1−δ
in

= Pq − wq δ = 0.
Pr

∂q δ
Thus, the supply function of the hospital is
δ
δPq
 
1−δ
q(Pq , w) =
w
6-u
Example Society with 2 (competitive) firms 1 and 2
and a good q.

lt
ira
q1(l) = l1/3

G
z-
q2(l) = l1/2

ne
ti
ar
Individual supply functions:

M
er
1
Pq
vi

2
Xa
q1(Pq , w) =
3w

c

Pq
s
ic

q2(Pq , w) =
om

2w
on

Aggregate supply:
Ec
lth

1 1/2
Pq Pq 2wPq + 3wPq

2
ea

q(Pq , w) = + =
2w(3w)1/2
H

3w 2w
of

Elasticities
es
pl

∂q1 Pq 1
ci

εq1 = =
in

∂Pq q1
Pr

2
∂q2 Pq
εq2 = =1
∂Pq q2

6-v
 Pq
Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
q1 (Pq , w)

s
c

Xa
vi
er
M
ar
q2 (Pq , w)

ti ne

q
z-
G
ira
q(Pq , w)

6-w
lt
Insurers

Private vs Public health care systems

lt
ira
Private market for health insurance

G
z-
ne
ti
- adjustment of premia to the individual risk: only

ar
M
weak solidarity

er
vi
Xa

- Efficiency
c
s
ic
om

* consumers can choose among a menu of poli-

on
Ec

cies
lth
ea
H

*insurers have incentives to control expenses

of
es
pl

- Equity
ci
in
Pr

* some individuals may not be insured (adverse

selection problem)

* different treatment of good and bad risks

7
 lt
ira
G
z-
ne
Readings:

ti
ar
M
er
- Setting priorities:
vi
Xa
Hitchen, L., 2006, Bid to cut waiting lists has pushed

c

safety down NHS agenda, BMJ 332(7537), Febru-

s
ic
om

ary 11: 324.

on
Ec

- Equity:
lth
ea

Deemong, C., and J. Keen, 2004, Choice and eq-

H

uity: lessons from long term care, BMJ 328(7453),

of
es

June 12: 1389-1390.

pl
ci
in
Pr

7-a
Public centralized system for health insurance

- compulsory insurance financed through taxes and/or

employer/employee contributions

lt
ira
- government regulation of the health care sector

G
z-
ne
- Efficiency

ti
ar
M
er
* limited choice for population
vi
Xa
* spending control through government policies

c
s
ic
om

- Equity
on
Ec

* universal coverage
lth
ea

* solidarity between good and bad risks

H
of
es

* other aspects of equity:

pl
ci
in

- equity of finance (cost-sharing by income; indiv.

Pr

election insurance public/private) vs. equity of ac-

cess (= treatment for = need; universal access) →
Deeming and Keen (2004).

- health care insurance → see ch. 7

7-b
3. The market and the health care market

“Place” where consumers and producers interact (i.e.

lt
ira
exchange goods).

G
z-
ne
What goods compose a market? → demand ori-

ti
ar
ented vs supply oriented

M
er
vi
Xa
Demand oriented: set of products with high crossed

c

elasticities among them and low wrt other goods.

s
ic
om
on

Examples
Ec

(a) crossed elasticity between 95 octane and 98 oc-

lth
ea

tane gasoline is high. They are close substitutes.

H

They belong to the same market.

of
es

(b) crossed elasticity between consumption of gaso-

pl
ci

line and mineral water is low. They are independent

in
Pr

goods. They belong to different markets.

PROBLEM: ambiguity of high/low enough crossed

elasticity.

8
Supply oriented:

- Europe NACE (General Industrial Classification of

Economic Activities [Nomenclature statistique des

lt
Activités économiques dans la Communauté Européenne]),

ira
G
z-
- Spain CNAE (Clasificación Nacional de Actividades

ne
ti
Económicas)

ar
M
er
- US NAICS (North American Industry Classification
vi
Xa
System)

c
s
ic

PROBLEM: codes assigned according to techno-

om

logically oriented criteria. May be misleading, e.g.

on

elaboration of wine and champagne have different

Ec

codes, but often grouped in the same market (high

lth
ea

crossed demand elasticity).

H
of

Imperative assumption in the study of a market:

es
pl

Rational behavior of agents:

ci
in
Pr

- consumers: maximize utility −→ individual demand

−→ Market demand

- firms: maximize profits −→ individual supply −→

Market supply
8-a
 Pr
in
ci
pl
es
of
H
ea

Market structures:
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G

8-b
ira
lt
PERFECTLY COMPETITIVE MARKET

lt
ira
G
z-
Justification:

ne
ti
ar
M
1. Simplicity.

er
vi
Xa

c
s

2. Generates the best allocation of resources (no

ic
om

mismanagement): efficient distribution (Pareto-

on

optimality) [6= equity].

Ec
lth
ea
H

3. No need of the State to achieve efficiency.

of
es
pl
ci
in

4. Benchmark to build models allowing better un-

Pr

derstanding of real phenomena.

8-c
Assumptions:

1. Many sellers (producers): price-takers; given

lt
ira
prices choose production volume to max profit.

G
z-
qj ∗
Q∗ =

ne
X
∞qj ∗, lim ∗ = 0;
j→∞ Q

ti
ar
j=1

M
qj∗(p, w) = argmaxq Π(q)

er
vi
Xa

c
s

2. Many buyers (consumers): price takers; given

ic
om

prices choose consumption bundle to max sat-

on

isfaction.
Ec
lth

x∗
∗ ∞xi∗, lim i∗ = 0;
X
x =
ea

i→∞ x
H

j=1
of

x∗i (p, m) = argmaxxU (x)s.t. budget constraint

es
pl
ci
in
Pr

3. Homogeneous product.

4. Perfect information.

8-d
 lt
ira
G
5. Free entry (and exit) of firms.

z-
ne
ti
ar
M
6. Partial equilibrium. Static set-up.

er
vi
Xa

c

Additional assumption:
s
ic
om
on

7. Real markets (no financial markets)

Ec
lth
ea

• markets of goods and services: firms sell;

H

consumers buy.
of
es
pl

• labor markets: firms buy; consumers sell.

ci
in
Pr

8-e
Implicit assumption: property rights

lt
ira
G
8. Firms (shareholders) hold the property right over

z-
ne
profits −→ incentives to reinvest to improve

ti
ar
profitability −→ ∆Π.

M
er
vi
Xa
9. Consumers hold the property rights over their

c
s

incomes:
ic
om
on

• incentives to work (increase income)

Ec
lth

• incentivos to save (increase returns of cap-

ea
H

ital)
of
es
pl

=⇒ ∆ consumption.
ci
in
Pr

A State setting incomes and profits eliminates in-

centives.

8-f
Incentives

Are necessary but ... generate inequality.

lt
ira
Induce proper behavior if linked to profitability: higher

G
profitability −→ higher income.

z-
ne
ti
ar
Consequence: trade-off between incentives and in-

M
er
equality.
vi
Xa

c

If society offers + incentives (e.g. ∇ Tx, ∇ social

s
ic

benefits) i.e. indiv. welfare. ∼ income

om
on


 ∆ production
Ec

−→
 ∆ inequality
lth
ea

If society offers - incentives (e.g. ∆ Tx, ∆ social

H
of

benefits) i.e. indiv. welfare depends of income and

es

social benefits
pl
ci


in

∇ production
Pr


−→
 ∇ inequality

Societies solve the trade-off between the two forces

through voting in government elections.
8-g
Prices

allocate goods and services through the market to

those with highest willingness to pay.

lt
ira
G
BUT is not the only allocation mechanism, e.g.

z-
ne
ti
(i) Rationing (the consumption bundles consumers

ar
M
get are smaller that what they wish)

er
vi
Xa

• por queuing (cinemas, primary care services,

c
s

...) −→ inefficient
ic
om
on
Ec

• por lotteries (licences, ...) −→ inefficient

lth
ea
H

• por sharing rules (prorate shares in privatization

of

of public firms, food stamp programs, wartime,

es
pl

...)
ci
in
Pr

- without market for coupons −→ inefficient

- with market for coupons −→ efficient

(ii) Fixing prices (electricity, house-rental, ....)

8-h
Market equilibrium: Law of demand and supply.

Aggregate demand and supply of a commodity x

jointly determine its (partial) equilibrium price (and

lt
ira
quantity) in a perfectly competitive market.

G
z-
ne
ti
ar
An equilibrium is a situation where no agent has in-

M
centives to modify his(her) actions.

er
vi
Xa

The equilibrium pair (P ∗, x∗) denotes a situation

c
s
ic

where firms are maximizing profits and consumers

om

are maximizing satisfaction from consumption.

on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

8-i
Ilustration

Recall the market demand in pp. 5j-5l and market

lt
ira
supply in pp. 6u-6w.

G
z-
ne
m
Demand :xD (Px, m) =

ti
ar
Px

M
1
P Px


er
S x 2
Supply :x (Px, w) = +
3w vi 2w
Xa

c

Assume m = 10 and w = 1/3, so that

s
ic

10
om

Demand :xD (Px) =

on

Px
Ec

S 1/2 3Px
Supply :x (Px) = Px +
lth

2
ea

Equilibrium is characterized by xD (Px) = xS (Px).

H
of

Formally,
es
pl

3Px 10
ci

1/2
Px ⇐⇒
in

+ =
Pr

2 Px
3 2
P + Px − 10 = 0
2 x
That is, Px ≈ 2.27 and x ≈ 4.40.

8-j
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G

8-k
ira
lt
Characterization of competitive equilibrium

- Firms (given prices) choose q to maximize profits,

lt
ira
Π(q) = pq − T C(q)

G
z-
∂Π(q) ∗ s.t. p = M C(q ∗ )
= 0 → q

ne
∂q

ti
ar
M
- free entry guarantees zero profits, Π(q ∗) = 0

er
vi
→ pq ∗ = T C(q ∗) → p = T C(q ∗)/q ∗ = AC(q ∗).
Xa

c
s

Hence, at q ∗, p = M C(q ∗) = AC(q ∗).

ic
om
on
Ec
lth
ea

M C(q) AC(q)
H
of
es
pl
ci
in
Pr

p MR

q∗ q

8-l
Equivalence Max profits and Min costs

- Profit maximization:
maxq Π(q) = pq − wL − rK s.t. q = f (K, L)

lt
ira
G
Isoprofit map: q = Π + wL + rK

z-
p p p

ne
→ optimum satisfies

ti
ar
w ∂f r ∂f

M
= , and = .

er
p ∂L p ∂K
vi
Xa

c

∂f
s

Thus, profits are maximized at w ∂L

r = ∂f .
ic
om

∂K
on
Ec

q K
lth

K given
ea
H
of
es

q = f (K, L)
pl

w
ci

α tg β = −
r
in

w K
Pr

tg α =
Π∗ p
p q∗

β
∗ ∗
L L L L
Profit maximization Cost minimization

8-m
- Cost minimization:

lt
ira
G
min wL + rK s.t. q = f (K, L)

z-
K,L

ne
ti
ar
M
Isocost map: K = TrC − w

er
rL
vi
Xa
→ optimum satisfies

c

∂f
s

w
ic

− = − ∂L .
om

r ∂f
on

∂K
Ec
lth

Conclusion:
ea
H
of

With given prices (p, r, w), max profits ⇔ min to-

es
pl

tal cost. If a firm max profits producing q ∗, it must

ci
in

be minimizing cost. Otherwise, it would mean there

Pr

is a cheaper way to produce q ∗ contradicting profit

maximization.

8-n
4. Regulation

Why does it exist a public sector?

lt
The State plays a double rol in the economy:

ira
G
z-
ne
- regulates the market (taxes, transfers, minimum

ti
wages, compulsory schooling, vaccination campaigns,

ar
M
...)

er
vi
Xa
- agent in the market −→ PUBLIC SECTOR (−→

c

Mixed Economy).
s
ic
om
on

Components of the Public Sector:

Ec
lth
ea

(a) Welfare State: Health care services (SS), Edu-

H

cation, Pensions, Defense (?).

of
es
pl
ci

(b) Services (Liberalization, Privatization): Railways,

in
Pr

Mail, Telecommunications, Airlines.

(c) Industry (Privatization): Mining, Energy, Iron and

Steel.

9
Characteristics of the Public Sector:

(i) its objective need not be profit maximization ;

lt
ira
G
(ii) managers of public firms are “reliable officials”;

z-
ne
ti
ar
M
(iii) State has the right to impose duties (e.g. taxes)

er
to citizens and self-imposes control mechanisms.
vi
Xa

c

The rol of the State in the Economy: Market failures

s
ic
om

and Intervention (Regulation).

on
Ec

If competitive markets are efficient, why is there any

lth
ea

need of State regulation?

H
of
es

Free competition raises problems, e.g. negative ex-

pl
ci

ternalities (pollution, insufficient education, ...) Also

in
Pr

there appear market failures −→ inefficiencies (ex-

clusion high risks individuals, ...), free-riding, so-
cial complaints, entry barriers (licence for activities:
banks, restaurants, physicians,...)

9-a
Mechanisms of regulation:

- direct (substituting the private sector);

public transport, health care provision, public education

lt
ira
- providing incentives to the private sector (price ma-

G
z-
nipulation via transfers/taxes);

ne
subsidies to private schools

ti
ar
M
- imposing rules to the private sector (legislation);

er
vi
min wage, age in labor market, safety workplace, antitrust laws
Xa

c
s

- combinations.
ic
om
on

Types of regulation:
Ec
lth

- universally accepted (access of kids to the labor

ea

market)
H
of
es

- controversial (positive action for gender/race)

pl
ci
in

- on producers/consumers (price discrimination; an-

Pr

titrust laws; controls on advertising; access of con-

sumers to info on products, ...)

- on production conditions (safety at workplace; patents;

waste disposal; environmental pollution, ...)
9-b
Reasons for regulation:

- protection of working conditions (health, safety, ...)

lt
ira
G
z-
- protection of vulnerable social groups (kids, immi-

ne
ti
grants, ...)

ar
M
er
vi
- protection of competitive conditions
Xa

c
s
ic

- prevention of market abuse

om
on
Ec

Instruments for regulation:

lth
ea
H

- laws (antitrust agencies)

of
es
pl

- administrative actions
ci
in
Pr

- professional associations (entry conditions

Objetive of the regulation: correct market failures.

9-c
Why does the State regulates the health care mar-
ket?

Typology of answers:

lt
ira
G
(a) the market is too complex for patients. Providers

z-
ne
would take advantage on them;

ti
(b) health is too a fundamental good for govern-

ar
M
ments let the market operate freely;

er
vi
(c) health care market generates externalities ( [-]
Xa
epidemies, [+] vaccinations);

c

(d) poor people must have access to the health care

s
ic
om

market;
on

(e) asymmetric information between physician and

Ec

patient (moral hazard, adverse selection).

lth
ea
H

BUT ... there are other markets for which these

of

arguments also apply and there is no intervention

es

(food, housing) neither in production nor in distribu-

pl
ci

tion.
in
Pr

ALSO,

- Health care market is not competitive → monopoly

power, public goods, externalities, ... (see below)
9-d
 lt
ira
G
z-
ne
- Hospitals

ti
ar
M
entry barriers (permissions, subsidies, ...)

er
size (scale and scope economies)
vi
Xa
asymmetric info hospital/patient (quality,...)

c

ownership (public, private, for profit, nonprofit)

s
ic
om
on

-Physicians
Ec

entry barriers
lth
ea

professional associations
H

private sector: price discrimination to patients and

of
es

insurers
pl
ci
in
Pr

9-e
 lt
ira
G
z-
ne
ti
ar
M
er
Reading: vi
Xa

c
s
ic

Philipson T., and E. Sun, 2008, Regulating the safety

om

and efficacy of prescription drugs, VoxEU.orp,

on
Ec
lth

[http://www.voxeu.org/index.php?q=node/804]
ea
H
of
es
pl
ci
in
Pr

9-f
SOURCES OF MARKET FAILURE

I. Supply side

(i) natural monopolies (scale economies) → large

lt
ira
initial investment: supply of water, gas, electricity,

G
z-
transport, telecommunications, ...

ne
ti
ar
Regulation (limit monopoly power) widely accepted

M
er
(prices)
vi
Xa
(ii) oligopolies (monopoly power) [see below]

c
s
ic
om

Regulation (limit monopoly power): antitrust laws

on
Ec

(iii) Externalities → difficult to measure, diversity of

lth

effects, diversity of types. [see below]

ea
H
of

Regulation (limit monopoly power): OK but how?

es
pl

(iv) public goods: no exclusion, no rivalry (public

ci
in

gardens, roads, army) [see Ch. 5]

Pr

Regulation (protect “monopoly rents”)

(v) Merit goods and incomplete markets [see be-

low]
10
 lt
ira
G
II. Demand side

z-
ne
ti
ar
(i) imperfect and incomplete information on products

M
er
(AIDS, drugs) and markets.
vi
Xa

c

Regulation: control on sales of dangerous products;

s
ic
om

info on label of products (expiry date, ingredients,

on

...); control on advertisement campaigns.

Ec
lth
ea

(ii) information as a public good → private market

H

does not provide enough information (see below).

of
es
pl
ci

Regulation: increase volume of information.

in
Pr

10-a
OLIGOPOLY

Consider a market with two firms (duopoly) 1 and 2.

Firm 1’s decision will be affected by firm 2’s behav-

lt
ira
ior → Strategic interaction

G
z-
ne
Pn
Now, Q = j=1 qj and qj /Q > 0!!

ti
ar
M
er
Firm 1’s decision-making process
vi
Xa

c

- Market price will depend on firms production lev-

s
ic
om

els: P (q1, q2). Therefore, demand downward slop-

on

ing.
Ec
lth
ea

1’s profit maximization: find production level solving

H
of
es

maxq1 Π(q1, q2) = q1P (q1, q2) − C(q1)

pl
ci
in
Pr

Solución: q1 = f (q2)

Similarly, firm 2 maximizes profits producing

q2 = g(q1)
11
 lt
ira
G
z-
Market equilibrium

ti ne
ar
M
(q1∗ , q2∗ ) such that f (q2) is compatible with g(q1)

er
vi
Xa

c

q1 P
s
ic
om

(MC=0)
q2 = g(q1 )
on

.Monopoly
Ec

q1m Pm
.
lth

Duopoly
Pd
q1∗
ea

q1 = f (q2 )
H

Demand
of

MR
es

q2∗ q2m q2 Qm Q∗ Q
pl

Q∗ = q1∗ + q2∗
ci
in
Pr

11-a
MONOPOLY

Profit maximization

maxq Π(q) = qP (q) − C(q) = I(q) − C(q)

lt
ira
G
z-
Marginal Revenue: ∆ revenue when selling one ad-

ne
ti
ditional unit

ar
M
er
Marginal Cost: ∆ cost when producing one addi-
vi
Xa
tional unit

c
s
ic

Average Cost: Total Cost/production (unit cost)

om
on

Firm’s problem: maxq Π(q), =⇒ M R = M C

Ec
lth

P
ea

MC
H
of
es

P∗
pl

Profits AC
ci
in
Pr

Cost Demand

MR
q∗ q

12
Monopoly power

Monopolist: pm > M C = pc ⇒ deadweightloss

lt
P

ira
G
z-
ne
Pm

ti
MC

ar
Pc

M
er
vi
Xa Demand

c
s

qc
ic

qm q
om

MR
on
Ec

Deadweightloss: Monopolist expels consumers un-

lth

able to pay P m → aggregate consumption ↓ (q c −

ea

q m) Remaining consumers pay higher price. Con-

H
of

sumer surplus ↓ upper yellow triangle.

es
pl
ci

Monopolist sells q m at higher price, but does not

in
Pr

produce (q c−q m) that could sell at a price > M C →

Producer surplus ↓ lower yellow triangle.
Note: no price discrimination

Overall loss of efficiency: yellow triangle.

12-a
Health care market contains elements potentially al-
lowing for market power:

- Hospitals with few competitors

lt
[rural vs. urban areas; specialities]

ira
G
z-
- Private hospitals w/ mkt. power

ne
[ad campaigns, contracts w/ insurers]

ti
ar
M
- Patented pharmaceutical products

er
vi
Xa
- Social Security (State monoposonist pharma mkt.)

c
s
ic
om

- Licences (exams) to enter the market

on
Ec

- Professional associations fixing minimum fees

lth
ea

i.e.
H
of
es

* Entry barriers induced or introduced by the gov-

pl

ernments to guarantee minimum quality standards,

ci
in

promote R&D, ...

Pr

* Government may decide to regulate non-profitable

situations: transfers donations to build a hospital
in a small community; offer nonprofitable services
(trauma, burnt, neonatal, intensive care units).
12-b
CRITICISM: regulation may worsen situation if not
adequate. BUT may improve situation if regulation
is efficient.

lt
Example: control on monopoly prices

ira
G
z-
P

ne
ti
ar
M
A
Pm

er
F

vi
P r MC
C
Xa
Pc

c

E
s

Demand
ic

B
om
on

qm qr qc q
Ec

MR
lth
ea

Monopolist hospital: (P m, q m) → welfare (dead-

H
of

weight) loss = ABC.

es
pl
ci

Government regulation: price cap P r → q r → wel-

in
Pr

fare (deadweight) loss = F EC < ABC.

Problem: hospital is multiproduct service provider

+ demand and technology evolve → difficult to reg-
ulate properly.
12-c
Measuring monopoly power

lt
♠ Firm level

ira
G
z-
ne
Lerner index: Li = Pi−M Ci
∈ [0, 1)

ti
Pi

ar
[i’s capacity to quote Pi above M Ci ]

M
er
vi
Xa
♠ Aggregate level: 3 measures

c

[rank firms from largest to smallest market share]

s
ic
om

Pk
i=1 Li
on

Lk =
k
Ec

[arithmetic mean of k largest firms]

lth
ea

n
q
H

miLi, mi = Pn i
X
La =
of

i=1 qi
es

i=1
pl

[arithmetic mean weighted by firms’ market shares]

ci
in

n
Pr

(L − I)mi , Li 6= 0
Y
Lg =
i=1
[geometric mean weighted by firms’ market shares]

12-d
 lt
ira
G
Example:

z-
ne
ti
ar
Rewrite Lerner index as (see below)

M
er
Pi − M Ci m
Li = =− ivi
Xa
Pi ε

c


s

P = 1.01M C
ic

i i
(i) Let εi = −100 (very elastic) ⇒
om

L = 0.01
i
on

Note 1: Pi ≈ M Ci →∼ compet behavior

Ec
lth
ea


P = 10M C
H

i i
(ii) Let εi = −10/9 (very low elast) ⇒
of

L = 9/10 = 0.9
i
es

Note 2: Pi ≫ M Ci →∼ monop behavior

pl
ci

Note 3: Recall P m always on elastic part of demand function

in
Pr

12-e
 lt
ira
G
z-
Example (cont): Let ε = −2; n = 5

ne
ti
ar
m1 m2 m3 m4 m5

M
0.4 0.25 0.2 0.1 0.05

er
L1 L2 L3 L4 L5
0.4 0.25 0.2 0.1 0.05

vi
− −2 = 0.2 − −2 = 0.125 − −2 = 0.1
Xa − −2 = 0.05 − −2 = 0.025

c
s
ic

3
1 X
om

Lk=3 = Li = 0.142
on

3 i=1
Ec

5
lth

X
La = miLi = 0.1375
ea

i=1
H
of

5
(Li)mi = 3.323
Y
Lg =
es
pl

i=1
ci
in
Pr

12-f
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G
ira

12-g
lt
 lt
ira
G
z-
ne
An illustration of market power:

ti
ar
M
er
Welfare effects of a switch from Rx to OT C
vi
Xa

c

Reading:
s
ic
om
on

Tina Shih,Y-C., M. Prasad, and B.R. Luce, 2002,

Ec

The effect on social welfare of a switch of second-

lth
ea

generation antihistamines from prescription to over-

H

the-counter status: A microeconomic analysis, Clin-

of
es

ical Therapeutics, 24: 701-716.

pl
ci
in
Pr

13
Welfare effects of a switch from Rx to OT C

1.- description of the set-up

lt
ira
G
z-
ne
2a.- Drugs close to patent expiration

ti
ar
M
2b.- Drugs under patent protection

er
vi
Xa

c

3.- Welfare effects of a switch from Rx to OT C. Re-

s
ic

sults:
om
on
Ec

- case a)
lth
ea

 
M CRx
 = M COT C 4W ambiguous
H

⇒
 
of

   
εRx 
  < εOT C 
 P
OT C < P Rx
es
pl
ci

- case b)
in
Pr

 
M CRx <M
 COT C
 4W ambiguous
⇒
 
   
εRx 
  < εOT C 
 4P ambiguous

13-a
Welfare effects of a switch from Rx to OT C

July ’88 Blue Cross California request to NDAC for

switching from Rx to OTC status 3 SGAs:

lt
ira
- fexofenadine hydrochloride [Allegra - Aventis]

G
z-
- loratadine [Claritin - Schering]

ne
- cetirizine hydrochloride [Zyrtec - Pfizer]

ti
ar
M
Regulation: An OTC product must show safe and

er
vi
effective when used without supervision of a health
Xa
care practitioner.

c
s
ic
om

Basic economic argument of request:

on

- cost savings due to expected price reduction in

Ec

SGAs after switch

lth

- based on observed pricing for H2-receptor antag-

ea

onists before and after switch to OTC status

H

BUT
of
es

H2-receptor antagonists switch was proposed by man-

pl

ufacturers near patent expiration

ci
in

WHILE
Pr

in SGAs case, all 3 drugs were atill under patent

protection.
THUS
- extrapolating conclusions → erroneous
- social welfare implications → uncertain.
13-b
[2a] Drugs close to patent expiration
Argument: after switch to OTC status, competition
will lower prices → higher consumer surplus (proxy
for social welfare)

lt
ira
G
z-
P

ne
ti
A

ar
M
er
B
vi
P0 Xa

c
s

P1 C
ic

D
om

demand
on
Ec
lth

Q0 Q1 Q
ea
H
of
es

CS(P0) = AP0B
pl

CS(P1) = AP1C = AP0B + P0P1DB + DBC

ci
in
Pr

4 CS = P0P1DB + DBC, that is,

4 CS from ↓ price + 4CS from ↑ consumption
Strategic switch to OTC status to preempt generic
competition.
13-c
[2b] Drugs under patent protection
Patent protection grants monopoly power to manu-
facturer → pricing rule: M R = M C ⇒ deadweight
loss wrt perfect competitive pricing.

lt
ira
G
P

z-
ne
A

ti
ar
M
er
Pm B

vi
Xa

c

C
Pc MC
D
s
ic

demand
om
on

MR
Ec

Qm Qc Q
lth
ea
H
of

CS(Pm) = APmB
es

CS(Pc) = APcC = APmB + PmPcDB + DBC

pl
ci
in

4 CS = PmPcDB + DBC, that is,

Pr

4 CS transfer to firms + 4CS deadweight loss

In addition to consider CS, discussions of welfare ef-
fects of switch from Rx to OTC must consider change
in deadweight loss before and after the switch.
13-d
[3.] Welfare effects of a Rx to OTC switch
2 questions
- will the price of SGAs ↓ after the switch?
- social welfare consequences of the switch?

lt
ira
Effects on demand

G
z-
- Health insurance → 4 demand (moral hazard)

ne
→ consumers less price sensitive

ti
ar
BUT no coverage for OTC drugs:

M
er
•Rx → OTC ⇒ sP → P (s: copayment rate)
• demand Rx less elastic than demand OTC vi
Xa

c
s
ic
om

P P
A
on

A!
Ec
lth

PRx
POT C
ea
H

M CRx = M COT C
of
es
pl

demand demand
ci
in

M RRx Q M ROT C Q
Pr

Rx market OTC market

(before switch) (after the switch)

Pricing strategy of a (profit max) patent holder?

assumption: M CRx = M COT C
13-e
- micro theory: εRx < εOT C → POT C < PRx
- BUT ↓ POT C no guarantee 4 welfare.
- Consider a drug under patent

lt
ira
P P

G
A

z-
A!

ne
ti
ar
PRx B

M
B!
POT C

er
vi
C M CRx = M COT C
Pc
Xa
E E !
C!

c

demand demand
s
ic
om

M RRx Q M ROT C Q
on

Rx market OTC market

Ec
lth
ea

Rx market OTC market

H
of

CS(PRx ) = APRx B CS(POT C ) = A! POT C B !

es

CS(Pc ) = APc C CS(Pc ) = A! Pc C !

pl
ci

!CS = Pc PRx BE + BEC !CS = Pc POT C B ! E ! + B ! E ! C !

in
Pr

WRx = APRx B + Pc PRx BE − BEC WOT C = A! POT C B ! + Pc POT C B ! E ! − B ! E ! C !

M N M! N!
!W = (M − M ! ) − (N − N ! ) >
< 0

sign4W depends on shape of demand curves → empirical

issue
13-f
Assumption M COT C > M CRx
- advertising
- new distribution channels
- new packaging, ...

lt
ira
THEN

G
z-
ne
POT C may be increased after switching to OTC sta-

ti
ar
tus

M
er
vi
Xa
P P

c

A
A!
s
ic

POT C
om

PRx M COT C > M CRx

on
Ec
lth

M CRx
ea
H

demand demand
of
es

M RRx Q M ROT C Q
pl

OTC market
ci

Rx market
in
Pr

THUS

not only welfare consequences are uncertain, but

the assumption of a post switch price reduction as
well.
13-g
Externalities

A good shows externalities when it generates third-

party effects outside the price system

lt
ira
G
- positive: vaccination of my neighbors on my chances

z-
ne
to get infected, etc.

ti
ar
M
- negative: pollution, neighbor’s loud music, etc.

er
vi
Xa
Competitive market only considers private costs and

c

benefits, not social ones → inefficiency:

s
ic
om

negative externalities → overproduction;

on

positive externalities → underproduction.

Ec
lth

Example: market of vaccination.

ea
H
of

D: demand (marginal private benefit)

es

S: supply (marginal private cost)

pl
ci

K: marginal external benefit

in
Pr

Initial situation: Competitive allocation A → ineffi-

cient under positive externality K:
marg. social benefit = D + K > S = marg. social
cost (= marg. private cost)
14
 lt
ira
Government intervention: direct subsidy to produc-

G
z-
0
ers of K e→ supply shifts to S = S − K.

ti ne
ar
M
New equilibrium allocation: q2 at price p2 → effi-

er
cient. vi
Xa

c

p
s
ic
om
on

A
p1
Ec

S
lth

p2 S =S−K
ea
H
of

D
es

K
pl
ci
in

q1 q2 q
Pr

14-a
Merit goods

lt
ira
Commodities that are “good” regardless of each in-

G
dividual’s preferences: arts, compulsory education,

z-
ne
compulsory social insurance, ...

ti
ar
M
er
Individuals because of info problems are not fully
vi
Xa
aware of the benefits obtained from their consump-

c

tion.
s
ic
om
on

Govt. regulation: promotion of their consumption.

Ec
lth
ea

Incomplete markets
H
of
es

Private insurers may not provide coverage for some

pl
ci

illnesses: AIDS, cancer, ...

in
Pr

Govt. regulation: public provision of insurance, com-

pulsory contracts on private insurers.

15
Regulating the health care market

Definition: Regulation consists in the implementa-

tion of NON-Market mechanisms to affect the quan-

lt
ira
tity, price, quality of a good exchanged in the market.

G
z-
ne
ti
Implementation: Governmental agency (Ministry of

ar
M
health, ...)

er
vi
Xa

Objective: guarantee a minimum level of quality, quan-

c
s
ic

tity, ... eliminating inefficiencies due to scale or scope

om

economies.
on
Ec
lth

Instruments:
ea
H
of

-Monetary:
es
pl
ci
in

*salaries to personnel in public outlets

Pr

* reimbursement conditions to hospital (per diem,

per admission, prospective, retrospective) by third-
party payers (public and private).
16
2 basic mechanisms:

lt
ira
G
(i) retrospective: → ex-post

z-
ne
ti
ar
M
(ii) prospective: payment rates fixed prior to the pe-

er
riod health care is provided → incentives for effi-
vi
Xa
ciency by limiting spurious spending. Problem: How

c

to calculate prices?[DRGs]
s
ic
om
on

-Quantity: indirect instruments. Need for approval if

Ec

budget deviations → limit excess supply of facilities,

lth
ea

promote higher use rates of existing facilities.

H
of
es

-Quality: Entry barriers → providers with minimum

pl
ci

standards (physicians, nurses, pharmaceuticals); con-

in
Pr

trol of admission protocols, minimum/maximum in-

hospital stay, adequacy of treatments, ...

16-a
-Other instruments:

* licensure laws, mandates, national drug agency,

universal and compulsory health insurance (public

lt
ira
or private)

G
z-
ne
ti
* Regulating insurance companies (∼ banks) →

ar
M
guarantee financial capacity

er
vi
Xa

* Fiscal waives on insurance contracts

c
s
ic
om

* Public health:
on
Ec
lth

+ info campaigns on alcohol, tobacco, drugs,

ea
H

working accidents, traffic, ...

of
es
pl

+prevention and control campaigns at schools,

ci
in

vaccination campaigns, ...

Pr

* Financing research in particular diseases (AIDS,

cancer, ...) in public (Universities) and private (labo-
ratories) research centers.
16-b
PPS and DRGs

DRGs instrument to determine rates of the PPS.

lt
ira
Def.: DRG attemps to represent a case type that

G
identifies patients with similar conditions and pro-

z-
ne
cesses of care.

ti
ar
M
er
Each DRG is given a flat payment rate calculated
vi
Xa
in part, on the basis of costs incurred for that DRG

c

nationally.
s
ic
om

Example
on
Ec
lth

Consider a community with n hospitals and let us fo-

ea

cus in hospital 1. It faces a demand D and M C(q) =

H
of

AC(q) = C0.
es
pl
ci

(a) Monopoly: provide q0 units of care at a price p0

in
Pr

(point M ).

(b) If hospital ∈ NHS subject to retrospective reim-

bursement: produce at point R, i.e. AC = p →
Π = 0.
16-c
(c) If hospital ∈ NHS subject to prospective reim-
bursement based on DRGs.
p

lt
ira
G
z-
p0 M

ti ne
ar
M
R

er
C0
A

vi
AC = C ∗ + AF C
Xa
p∗ C∗

c

B
C1
s
ic
om

Lump-sum
on

transfer
Incremental
Ec

profit D
q1
lth

q0 MR q
ea
H

- Assume 1 sickness
of
es

- Assume hospital’s effort to reduce cost is a fixed

pl
ci

cost.
in
Pr

- Rates are fixed at the average marginal cost of

competitors (2, . . . , n) = C ∗
- Also fixed cost is reimbursed as lump-sum transfer.
- Thus, “new” average cost AC = C ∗+AF C(average
fixed cost)
16-d
 lt
Equilibrium: point B where C ∗ = D, i.e. (q1, p∗) so

ira
G
that Π = 0 because p = AC and f ixedcost0transf er.

z-
ne
ti
ar
J
Assume cost reduction yields “new” marginal cost

M
er
= C ∗ → hospital zero (incremental) profits.
vi
Xa

c

J
Assume cost reduction yields “new” marginal cost
s
ic

= C1 < C ∗ → hospital (incremental) profits =

om
on

q1(C ∗ − C1).
Ec
lth
ea

J
Assume cost reduction yields “new” marginal cost
H

= C ∈ (C ∗, C0) → hospital loses.

of
es
pl
ci

Conclusion: As hospital’s costs do not enter into the

in
Pr

price it receives, hospital big incentive to lower cost.

16-e
5. Public goods

lt
ira
G
* Public goods vs Public provision of (private) goods.

z-
ne
ti
ar
↓ ↓

M
er
vi
Xa
no rivalry because of natural monopoly

c
s
ic
om

no exclusivity subsidize particular users

on
Ec

Examples:
lth
ea
H

PG: free highways, gardens, bridges, roads, national

of
es

army, ...
pl
ci
in
Pr

PPG: private highways (tolls), mail, garbage collec-

tion, public transport, ... [exclusion if no pay]

17
* Why does market fail?

Private market underprovides public goods → “free

lt
riders” [→ no exclusion → no incentives to finance

ira
G
(individual impact negligible)]

z-
ne
ti
ar
Examples:

M
er
vi
Xa
- lack of prevention at work because of accident in-

c

surance
s
ic
om
on

- “free ride” in the bus

Ec
lth
ea

- inhabitants of BCN using public gardens in Sant

H

Cugat
of
es
pl
ci

- avoid vaccination when everybody gets vaccinated

in
Pr

(The Economist 11-4-98)

Conclusion: underprovision by private markets →

government regulation.

17-a
Are health care services public goods? NO!

- there is rivalry (services to a patient cannot be si-

multaneously provided to another)

lt
ira
G
- there may be exclusion (if patient cannot pay, may

z-
ne
be excluded)

ti
ar
M
⇒ Health care services: private good publicly pro-

er
vi
vided Xa

c

BUT
s
ic
om
on

INFORMATION is an economic good with a high

Ec

component of public good:

lth
ea
H

- getting info by one individual does not limit possi-

of

bilities to ohers.
es
pl
ci

- although information has a price, the cost of pro-

in
Pr

viding information to an additional individual is neg-

ligible (advertising campaigns, mailings, ...)

⇒ private market underprovides information → pub-

lic provision. How to do it?
17-b
- help disseminating info (public provision or trans-
fers to private provision)

lt
ira
- increase information participating in scientific re-

G
z-
search (public provision or transfers to private provi-

ne
ti
sion)

ar
M
er
ALSO vi
Xa

c
s
ic

Donations (charity) have characteristic of a public

om

good: help people with low income levels to improve

on
Ec

consumption (also health care)

lth
ea

Individual donations are small in the aggregate →

H
of

free riding [donation to help victims of natural disas-

es
pl

ters ...]
ci
in
Pr

How to implement these donations? (i) additional

income free to use; (ii) transfer with a predefined
use → problems of equity, efficiency, need, ...

17-c
 lt
ira
G
z-
ne
ti
ar
Reading

M
er
vi
Xa
Clinical trials as a public good:

c
s
ic
om

Lewis, T.R., J.H. Reichman, and A.D. So, 2007, The

on

case for public funding and public oversight of clini-

Ec

cal trials, Economists’ Voice, 4: article 3.

lth
ea
H

http://www.bepress.com/ev/vol4/iss1/art3/
of
es
pl
ci
in
Pr

17-d
 lt
ira
G
6. Nonprofit Enterprises/Organizations (NPE)

z-
ne
ti
ar
NPEs are important in the health care market (hos-

M
er
pitals and nursing homes):
vi
Xa

c

Definition: A NPE is an organization where nobody

s
ic
om

holds property rights on profits =⇒

on
Ec

- allows for objectives 6= max. profits,

lth
ea
H

- exempt from corporate taxes (profits, VAT, ...)

of
es
pl
ci

- donations to NPEs receive special fiscal treatment

in
Pr

18
Why do NPE exist?

2 reasons: (i) market inefficiencies, (ii) asymmetric

lt
information

ira
G
z-
ne
(i) Market inefficiencies (not solved by governmental

ti
ar
regulation)

M
er
vi
Xa
3 types of firms in a market:

c
s
ic
om

- for profit private firms,

on
Ec

- public firms,
lth
ea
H

- Nonprofit firms (public and private).

of
es
pl
ci

Public firms appear to correct market failures (exter-

in
Pr

nalities, public goods) but no guarantee of efficiency.

* Externalities → private market yields underpro-

vision → inefficiency. Why?

18-a
Example 1: vaccination for influenza.

lt
ira
G
- Individual ↓ fall sick (private benefit)

z-
ne
ti
ar
- Individual ↓ infect others (social benefit): External-

M
er
ity.
vi
Xa

c

Free markets: individual only computes private ben-

s
ic
om

efits → ignore social benefits → market delivers quan-

on

tity lower than max social benefit → inefficiency.

Ec
lth
ea

Government reaction: public firm to provide a public

H

good → avoid free rider problem.

of
es
pl
ci

BUT does not guarantee efficiency (potential inef-

in
Pr

ficiencies of governmental activities may offset po-

tential gains)

18-b
Example 2: campaigning for provision of a public
good.

lt
Society with 5 individuals: common marginal income

ira
G
tax (MIT).

z-
ne
ti
ar
Government proposes building a hospital for chil-

M
er
dren. Question: how big (how many beds)? VOT-
vi
Xa
ING (majority rule)

c
s
ic

Individual demands Di, (i = 1, 2, 3, 4, 5)

om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

Voting outcome for the different alternatives:

18-c
 lt
ira
G
z-
ne
ti
ar
M
er
vi
Xa

c
s
ic
om

Government provides C beds, every individual will

on

pay c × M IT and fiscal revenues = cost of the

Ec
lth

project. (Individuals 4,5 vote for C as a “second

ea

best” alternative).
H
of
es

Conclusion: Voting only satisfies individual 3 for whom

pl
ci
in

M IT = marginal benefit.
Pr

Consequence: government inefficient in providing

the public good → individuals 4,5 form a NPE and
provide the rest of the public good.

18-d
 lt
ira
Application to the health care sector.

G
z-
ne
ti
Health care is a private good (exclusion, rivalry) with

ar
M
externalities.

er
vi
Xa

Hospitals and nursing homes → room for NPE when

c
s
ic

there is enough dissatisfaction with public provision

om

↓
on
Ec
lth

- non-governmental organizations (greenpeace,

ea

medicus mundi, xxx sans frontières, ...)

H
of
es
pl

- origin of hospitals as charitable institutions re-

ci
in

lying on donations.
Pr

18-e
 lt
ira
G
z-
ne
ti
(ii) Asymmetric information

ar
M
er
vi
Different argument: difficulty in writing a complete
Xa

contract because quantity or quality are not (per-

c
s
ic

fectly) observable by the purchaser.

om
on
Ec

Example: assistance 3rd world.

lth
ea
H

Too costly to verify that a for-profit firm delivers goods

of

to targeted population → delegate in a NPE.

es
pl
ci
in
Pr

18-f
Example: nursing home

Similar to hotel management (rooms, housekeep-

ing, meals, recreation facilities)

lt
ira
G
z-
Differences

ne
ti
ar
M
hotel max profit s.t. satisfaction of clients

er
vi
Xa
nursing home clients (patients) are not able to eval-

c

uate quality of the service given their health state.

s
ic
om

Relatives only partial info.

on
Ec

* If nursing home is for-profit, may arise conflict

lth
ea

of interest with some clients.

H
of
es

* Empirical evidence: quality of services non-

pl
ci

profit nursing homes not inferior to for-profit nursing

in
Pr

homes.

Comment Not easy to reach consensus on quality

criteria → proxies: observable signals (variation in
health state of inmates, ...)
18-g
What is a hospital

lt
ira
G
1.- What does a hospital do?

z-
ti ne
ar
M
er
PCC/OPV

vi
Xa

c

Patient Initial
health state H (treatment) Final health
( discharge/
death )
s

state
ic
om
on

Emergency
Ec
lth
ea

2.- Hospital’s idiosyncrasy

H
of
es

• personalized treatment to every patient

pl
ci

• detailing → distort treatment selection

in
Pr

• moonlighting → distort allocation effort/quality

• resource hoarding

19
3.- Measures of activity

♠ # patients treated:

lt
ira
G
z-
- aggregate

ne
ti
ar
M
- per DRGs

er
vi
Xa

- per discharges/out-patient visits/emergency room

c
s
ic

episodes
om
on
Ec

♠ mortality rate
lth
ea
H

- Summary
of
es
pl

Hospital unit of multiproduct services →

ci
in
Pr

 Scale economies

 Scope economies

19-a
Modeling a hospital as a NPE

Three initial lines of research:

lt
ira
(a) Quantity vs. quality: Newhouse (1970)

G
z-
max U (Q, q) s.t. Π = 0

ne
Q,q

ti
ar
(Q, q): (Quantity, quality) of health care services.

M
er
vi
Xa
(b) As a cooperative Pauly-Redisch (1973)

c

net hosp. revenues

s
ic

max
om

M #physicians in coop.
on

HR ∼ Π. Closed/open cooperative.
Ec
lth
ea

(a), (b) comparable

H
of

(c) NPE vs. FPE Harris (1976)

es
pl
ci
in

Efficiency NPE vs. efficiency FPE, via property rights

Pr

basic argument:
manager reimbursement: trade-off between pecuniary & non-
pecuniary components
NPE: ↓ salary, ↑ NPI than FPE
20
Modeling a Nonprofit hospital (I): Newhouse (1970)

Organization of a hospital as a NPE: 3 decision-

making parties:

lt
ira
- Board of trustees → patrons 6= profiles,

G
z-
ne
- Hospital administrator (CEO) → agent of trustees,

ti
ar
M
er
- Physician staff.
vi
Xa

c

Assumption 1: complex decision-making → single

s
ic

utility function.
om
on

Assumption 2: Utility function U (Q, q) defined over

Ec

single sickness.
lth
ea
H

Q quantity of output (cases treated/time period);

of
es
pl

q quality of output: index aggregating physician

ci
in

staff expertise, hospital facilities, nursing care to pa-

Pr

tients, ...

Objective of the hospital: select (Q, q) to max U (Q, q)

given budget constraint (revenue form donations and
patients = cost).
20-a
Demand of health services depend on quality D(qi).

lt
ira
G
z-
ne
ti
ar
M
er
vi
Xa

c
s
ic
om

Costs of hospital depend on quality. Average cost

on

AC(qi)
Ec
lth
ea
H
of
es
pl
ci
in
Pr

20-b
Approach 1 (Equilibrium): given a quality level q1,

lt
ira
budget constraint (i.e. zero-profit condition) requires

G
z-
demand = average cost ⇒ Q1

ne
ti
ar
M
er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of

Note: D = AC ⇒ Π = 0
es
pl
ci

C(Q)
in

P (Q) = ⇔ P (Q)Q = C(Q) ⇔ Q(Q)Q−C(Q) = 0 = Π

Pr

20-c
ALERT: Quality and quantity non-monotonic rela-
tions:

If patients evaluate an 4 in quality q1 → q2 more

lt
ira
than an 4 increase in cost, we obtain Q1 < Q2, i.e.

G
z-
4q → 4Q.

ne
ti
ar
M
If average cost of providing a high level of quality q3

er
vi
is higher than consumers willingness to pay D(q3),
Xa
we obtain Q3 < Q2, i.e. 4q → 5Q.

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

20-d
Approach 2 (Equilibrium): find (Q, q) that max U (Q, q)
given budget constraint, i.e. Π = 0.

Equivalently, find Q − q frontier (feasible set) and

lt
combine with utility levels (indifference map).

ira
G
z-
ne
ti
ar
M
er
vi
Xa

c
s
ic
om
on
Ec

Hospital problem:
lth
ea

max U (Q, q) t.q. (Q, q) ∈ frontier Q − q

H
of
es
pl
ci
in
Pr

20-e
Modeling a Nonprofit hospital (II): Pauly-Redisch (1973)

Hospital as a physicians’ cooperative to maximize

their net income.

lt
ira
G
z-
Assume revenues only from bills to patients (no do-

ne
nations) → price of care determined by demand.

ti
ar
M
er
Supply of health care services depends on quantity
vi
Xa
of inputs: capital K, (non-technical) labor L, and

c

physicians M .
s
ic
om
on

Objective of the hospital: max net revenue per physi-

Ec

cian max HRM , where

lth
ea
H

HR = revenues − payments to (K, L, M ).

of
es
pl

Let M ∗ denote the optima number of physicians closed

ci
in

staff cooperative.
Pr

Given (K, L)

HR ↑ if M < M ∗ , and HR ↓ if M > M ∗ .

M M
21
 lt
ira
G
z-
ne
ti
ar
M
er
vi
Xa

c
s
ic
om

If open staff cooperative and competitive physicians

on
Ec

market →
lth
ea
H

physicians supply perfectly elastic S → optimal

of

number of physicians Ms.

es
pl
ci
in

In any case, contracting of (K, L) according to M ∗

Pr

or Ms.

21-a
Newhouse vs Pauly-Redisch

Simplifying assumptions

lt
ira
G
* Combine hospital and physicians revenues in a

z-
ne
single function R(Q, q)

ti
ar
M
er
* (Q, q) depend of (K, L, M )
vi
Xa

c

* Other hospital revenues: donations D + transfers

s
ic
om

(subsidies) G
on
Ec

* Perfectly competitive physicians market: price s

lth
ea
H

* Perfectly competitive capital market: price r

of
es
pl
ci

* Perfectly competitive labor market: price w

in
Pr

Hence,

HR = R(Q, q) + D + G − (wL + rK + sM )

22
Pauly-Redisch: max HR → HR appropriated by
cooperative.

lt
ira
Newhouse: max U (Q, q) s.t. HR = 0 Solution: A

G
z-
ne
ti
ar
M
er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es

Different outcome because cooperative when ap-

pl
ci

propriating HR includes D and G. It is “as if” hos-

in
Pr

pital would be a cover for a for-profit organization →

Policy conclusion: eliminate G and corporate tax ex-

emptions.

22-a
Comment: Is it possible Pauly-Redisch → Newhouse?

lt
ira
G
z-
YES, under some conditions, e.g.

ne
ti
ar
M
4 competition among hospitals (free entry). If de-

er
mand ∼ constant, HR ↓. Limit case HR → 0.
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

22-b
Modeling a Nonprofit hospital (III): Harris (1976)

Aim: compare efficiency NPE vs FPE → theory of

lt
ira
property rights

G
z-
ne
ti
ar
The theory of property rights

M
er
vi
Private firm: owner holds property rights on net prof-
Xa

c

its → may sell those rights (shares, ...)

s
ic
om

Large private firms: property 6= management. Man-

on
Ec

agers are agents of owners BUT imperfect agents

lth

because,
ea
H
of

* Cost of monitoring the manager → owner ac-

es
pl

cepts deviations from max Π.

ci
in
Pr

* in particular, manager also obtains “non pecu-

niary income” (NPI) → manager’s utility depends of
Π and NPI .../...

23
 lt
ira
.../... in turn, NPI will affect wealth generated by firm

G
z-
(profits). Hence, trade-off (Π, NPI)

ne
ti
ar
M
er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of

Example Hospital management opens a heart surgery

es
pl

outlet, not profitable but prestige → economic result

ci
in

(profits) will be affected.

Pr

23-a
 lt
ira
G
Property rights and for-profit enterprises

z-
ti ne
ar
FPE: selects feasible (Π, N P I) to max U (Π, N P I).

M
er
vi
Xa
Solution: A

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

23-b
Property rights and nonprofit enterprises

NPE’s problem: set manager’s wage to avoid appro-

priation of profits → 2 restrictions:

lt
ira
- feasible (Π, N P I) [1]

G
z-
ne
- max salary: L [2]

ti
ar
M
er
NPE: selects (Π, N P I) to max U (Π, N P I) s.t. [1]
and [2] vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

Solution: lower salary and higher NPI.

If NPI involve excessive non productive activities,

NPEs less efficient than FPEs.
23-c
Conclusion

lt
ira
G
- (III) is a different model because hospital does not

z-
ne
max profits as managers are imperfect agents.

ti
ar
M
er
- NPE no access to incentives on profits but in mod-
vi
Xa
els (I) and (II)hospitals take efficient decisions.

c
s
ic
om

- more adequate model? → empirical evidence:

on
Ec

* no significant differences between nonprofit and

lth
ea

for-profit hospitals ⇒ models (I) and (II) better than

H

(III).
of
es
pl
ci

* significant differences between nonprofit and

in
Pr

for-profit nursing homes ⇒ model (III) better than

(I) or (II).

23-d
Lines of development of literature

lt
ira
G
z-
♣ 1 illness → several illnesses

ne
ti
ar
M
♣ random demands (epidemic events, accidents,

er
...) vi
Xa

c
s
ic

♣ emergency vs. regular health care services

om
on
Ec

♣ primary health care integrated/segregated hospi-

lth

tal care
ea
H
of

♣ estimation of hospitals’ PPF: technical efficiency,

es
pl

...
ci
in
Pr

♣ etc.

23-e
 lt
ira
G
z-
ne
ti
ar
M
er
Reading
vi
Xa

c

Eggleston, K., Y-C. Shen, J. Lau, C.H.Schmid, and

s
ic

J. Chan, 2008, Hospital ownership and quality of

om
on

care: what explains the different results in the lit-

Ec

erature, Health Economics, 17: 1345-1362.

lth
ea

http://www3.interscience.wiley.com/journal/121519547/issue
H
of
es
pl
ci
in
Pr

23-f
7. Preventive vs. Curative Medicine. A health
policy exercise

The general set-up. Identify tools to incentivate:

lt
ira
G
- Patients to develop healthy habits

z-
ne
- Insurance companies, to provide preventive medicine

ti
ar
coverage.

M
er
Initial scenario:
vi
Xa

c
s

- Insurer covers hospital and treatment expenses.

ic
om

- Patient losses salary and bears pain.

on
Ec

⇒ split responsibility between insurer and insuree.

lth
ea
H

⇓
of
es
pl

Insurance contracts do not cover both demand sources

ci
in

→ underinvestment in preventive medicine.

Pr

Socially, investment in preventive medicine is prof-

itable. BUT insufficient individual incentives.
24
 lt
ira
Example [Helwege, 1996]

G
z-
ne
ti
Thesis: Policy misdesign because ignores some in-

ar
M
teractions among agents (patients, insurer)

er
vi
Xa

Compute costs of preventive vs. curative measures

c
s
ic

in men with high risk of heart attack.

om
on
Ec

Preventive measures:
lth

- Weight control
ea

- Yearly check-ups and cholesterol control.

H
of
es
pl

Curative measure:
ci
in

- Hospitalization and treatment of heart attack.

Pr

24-a
Data:
1. Population: Men, 50 years old (today).
?15% will suffer a heart attack before age 65;
? average age of heart attack: 60.

lt
ira
G
2. Cost of treatment: 30,000 e/individual

z-
ne
ti
ar
3. Cost of weight control: 50 e/individual/year

M
er
vi
4. Cost of check-up: 200 e/individual/year
Xa

c
s
ic

5. Incidence of weight control on risk: 5%

om
on
Ec

6. Incidence of weight control on cost:

lth

(0.05)(30,000)=1,500 e/year (for 10 years).

ea
H
of

7. Incidence of check-up on risk: 20%

es
pl
ci

8. Incidence of check-up on cost:

in
Pr

(0.2)(30,000)= 6,000 e/year (for 10 years).

9. Interest rate: 5%

10. Individual’s loss due to heart attack: 100.000 e

24-b
Discounted present value (time value of money)

[money today is more valuable than money in the future by the

amount of interest that money can earn.]

lt
ira
G
Compute the investment today at a return rate of

z-
ne
5%, so that in three years we will receive 100 EUR:

ti
ar
x(1.05)3 = 100

M
er
100
vi
x= 3
= 86.383759
Xa
1.05

c

In general, the present value of a capital K to be

s
ic

available in n years at the interest rate r is given by,

om
on

K
x=
Ec

(1 + r)n
lth

Example:
ea

- 1EUR at r = 5% in 10 years: 1.0510 = 1.629

H
of
es
pl

- present value in 10 years of weight control:

ci
in

50
Pr

= 30.69
1.629
- present value in 10 years of check-up:
200
= 122.77
1.629
24-c
Discounted present value: two approaches

lt
ira
(i) Value of 100 e today in 3 years discounted at 5%

G
z-
ne
t = 0 : 100

ti
ar
t = 1 : 100 + 5%(100) = 100 + 5 = 105

M
er
t = 2 : 105 + 5%(105) = 105 + 5.25 = 110.25
vi
Xa
t = 3 : 110.25 + 5%(110.25) = 110.25 + 5.5125 =

c

115.7625
s
ic
om

Summarizing: y = 100(1.05)3 = 115.7625 e

on
Ec
lth

(ii) Capital to invest to day at 5% interest rate so that

ea

in three years it is 100 e

H
of

x(1.05)3 = 100
es
pl

100
ci

x= = 86.38
in

3
Pr

(1.05)
That is investing today 86.38 eat 5%, yields 100 ein
three years.

24-d
Costs of performing weight controls and check-ups
along 15 years

lt
ira
G
z-
ne
Year Discount Weight control Check-up

ti
ar
0 0 50 200

M
1 1.050 47.62 190.48 200

er
1.05
2 1.103 45.63 vi 181.32 200
Xa
(1.05)2
3 1.158 43.18 172.71 200

c

(1.05)3
s

4 1.216 41.12 164.47 200

ic

(1.05)4
om

5 1.276 39.18 156.74 200

(1.05)5
on

6 1.340 37.31 149.25 200

(1.05)6
Ec

7 1.407 35.54 142.15 200

(1.05)7
lth

8 1.477 33.85 135.41 200

ea

(1.05)8
9 1.551 32.24 128.95
H

200
(1.05)9
of

10 1.629 30.69 122.77 200

(1.05)10
es

11 1.710 29.24 116.96 200

pl

(1.05)11
ci

12 1.796 27.84 111.36 200

in

(1.05)12
Pr

13 1.886 26.51 106.04 200

(1.05)13
14 1.980 25.25 101.01 200
(1.05)14
Total 544.90 2179.62

24-e
Insurer’s incentives to invest in prevention in one in-
dividual

? Expected cost of a heart attack (to happen in 10

years):

lt
ira
(0.15)(30000) = 4500 e

G
z-
ne
Discounted present value of this expected cost:

ti
ar
4500 4500

M
= = 2762.43 e

er
1.0510 1.629
vi
Xa
? Expected cost savings from weight control (5%):

c

(0.05)(2762.43) = 138.12 e
s
ic
om

? Expected cost savings from check-ups (20%):

on

(0.2)(2762.43) = 552.48 e
Ec
lth

? Discounted present value of cost weight control

ea

along 15 years: 544.90 e

H
of

? Discounted present value of cost check-ups along

es

15 years: 2179.62 e
pl
ci
in

Conclusion
Pr

138.12 < 544.90

552.48 < 2179.62
Insurer does not provide contracts with coverage for
preventive services. [neither separately nor together]
24-f
Patient’s incentives to invest in prevention

Patient’s loss: 100000 e(wage, pain, ...)

? Expected loss from a heart attack (to happen in
10 years time):

lt
ira
G
(0.15)(100000) = 15000 e

z-
ne
Discounted present value of this expected loss:

ti
ar
M
(15000) 15000
= = 9208.10 e

er
1.05 10 1.629
vi
Xa
? Expected cost savings from weight control (5%):

c

(0.05)(9208.10) = 460.40 e
s
ic
om

? Expected cost saving from check-ups (20%):

on

(0.2)(9208.10) = 1841.62 e
Ec
lth

? Discounted present value of cost weight control

ea

along 15 years: 544.90 e

H
of

? Discounted present value of cost check-ups along

es

15 years: 2179.62 e
pl
ci
in

Conclusion
Pr

460.40 < 544.90

1841.62 < 2179.62
Patient does not invest in preventive medicine (healthy
habits). [neither separately nor together]
24-g
Joint patient-insurer incentives to invest in preven-
tion

lt
ira
G
? Expected cost savings from weight control:

z-
ne
138.12 + 460.40 = 598.52 e

ti
ar
M
? Expected cost savings from check-ups:

er
vi
Xa
552.48 + 1841.62 = 2394.10 e

c
s

? Discounted present value of cost weight control

ic
om

along 15 years: 544.90 e

on

? Discounted present value of cost check-ups along

Ec

15 years: 2179.62 e
lth
ea
H
of

Conclusion
es
pl

598.52 > 544.90

ci
in
Pr

2394.10 > 2179.62

The alignment of incentives of insurer AND patient
makes investment in prevention profitable.

24-h
Comments

(i) Technological change increases uncertainty in

computation of costs. Compare

lt
ira
5 treatment costs

G
> 1 ⇒ prevention less attractive

z-
5 preventive costs

ne
ti
ar
M
er
(ii) Valuation of risk and benefits of prevention.
vi
F Common objection: psychological costs> hos-
Xa

pital costs+ loss of income. Maybe, but not for

c
s

everybody.
ic
om

F Willingness to pay vs. willingness to accept

on
Ec

compensation. Common problem in cost-benefit

lth

analysis.
ea
H

Example 1. Money parents willing to pay to pre-

of

vent death of an offspring 6= money they would

es
pl

accept as compensation after death of that off-

ci
in

spring.
Pr

Example 2. Patients that cannot afford paying

for drugs to prevent cholesterol, need not be
willing to accept compensation of 30000 e to
have a heart attack.

25
 lt
ira
(iii) Patients’ information

G
Imperfect info on:

z-
ne
F expected loss from a health crisis;

ti
ar
F risk of suffering a health crisis;

M
er
F benefits of prevention.
vi
Xa
⇒ info campaigns to induce healthy habits.

c
s
ic
om
on

(iv) Patients’ risk aversion.

Ec
lth
ea
H

(v) Preventive free services:

of

F physical exercise;
es
pl

F use of seat belts in cars.

ci
in

I Why not? Difficult answer: psychology.

Pr

25-a
How to solve inefficiency?

lt
ira
(i) Agreements insurer/insured: 5 premiums if in-

G
sured also contracts preventive services.

z-
ne
ti
ar
 insurer must invest in monitoring

M
 compare (monitoring cost + 5 revenues) with

er
hospital cost savings. vi
Xa

c

BUT
s
ic
om

F variety of illnesses and risk attitudes → large

on
Ec

menu of discounts (hard to manage)

lth
ea

F high risk groups may not be able to control

H

all elements of risk.

of
es
pl

F preventive treatments (with medicines) may

ci
in

have side effects (trigger other illnesses).

Pr

F low income patients may not have access to

discounts in premiums.

26
 lt
ira
(ii) Government subsidies to insurer and insuree to

G
incentivate prevention.

z-
ti ne
BUT

ar
M
er
F best use of tax revenues?
vi
Xa

F may be justified for public goods. Health

c
s

Care services are not.

ic
om
on

F tool to redistribute income. consequences:

Ec
lth

X who benefits from it?

ea
H

X who finance it (via taxes))?

of
es
pl
ci
in

(iii) General information campaigns.

Pr

26-a
Readings

- on incentives in healthy habits:

lt
ira
Helwege, A., 1996, Preventative versus Curative Medicine:

G
z-
A Policy Exercise for the Classroom, J. of Economic

ne
ti
Education, 27: 59-71.

ar
M
er
- on preventive care: vi
Xa

c
s
ic

Hennesy, D.A., 2008, Prevention and cure efforts

om

both substitute and complement, Health Economics,

on
Ec

17: 503-511.
lth
ea

http://www3.interscience.wiley.com/journal/117932700/issue
H
of
es

- on prenatal care:
pl
ci
in
Pr

Conway, K.S., and A. Kutinova, 2006, Maternal health:

does prenatal care make a difference?, Health Eco-
nomics, 15: 461-488.

http://www3.interscience.wiley.com/journal/112589788/issue

26-b
8. Uncertainty, Risk and Insurance

Individual: income Y , Utility U (Y ).

lt
Two states: healthy, sick (prob. p) → Yh, Ys

ira
G
z-
ne
• Expected income: ex-ante average income weighted

ti
by p: E(Y ) = pYs + (1 − p)Yh

ar
M
• Utility of expected income: U (E(Y ))

er
vi
• Expected utility: ex-ante average utility weighted
Xa
by p: E(U ) = pU (Ys) + (1 − p)U (Yh)

c
s
ic
om

Ilustración
on


p = 1/3
Ec



1 (0) + 2 (150) = 100
lth

Y s = 0 ⇒ E(Y ) = 3 3
ea


Yh = 150


H
of
es

1 1
U (Y ) = Y 2 ⇒ U (E(Y )) = 100 2 = 10
pl
ci
in


Pr

U (Y ) = U (0) = 0
s
U (Y ) = U (150) = 12.25
h

1 2
E(U ) = (0) + (12.25) = 8.16
3 3
27
Crucial elements of the analysis:

lt
ira
(a) E(U ) vs. U (E(Y ))

G
z-
ne
(b) Uncertainty vs. risk

ti
ar
M
er
Individual behavior facing probability of illness?
vi
Xa

c

Distinguish Uncertainty and Risk

s
ic
om
on

Def.: Risk
Ec
lth
ea

Individual can assign probabilities to the different

H
of

states he may face.

es
pl
ci

Def.: Uncertainty
in
Pr

Probabilities of the different situations are exoge-

nous.

27-a
Example 1 (Risk): 2 situations

1. Careless driver:

lt
ira
Prob 1/10,000 → accident

G
z-
ne
ti
Prob 9999/10,000 → no accident

ar
M
er
2. Careful driver: vi
Xa

c
s
ic

Prob 1/100,000 → accident

om
on
Ec

Prob 99,999/100,000 → no accident

lth
ea
H

Remarks
of
es
pl

1. Driving style is a choice variable.

ci
in
Pr

2. Occurrence of accident no proof of careless driv-

ing.

3. Occurrence of acccident is observable.

27-b
Example 2 (Uncertainty): 2 situations

1. Able physician:

lt
ira
Prob 1/100 → wrong diagnose

G
z-
ne
ti
Prob 99/100 → correct diagnose

ar
M
er
2. Lesss-able physician: vi
Xa

c
s
ic

Prob 2/100 → wrong diagnose

om
on
Ec

Prob 98/100 → correct diagnose

lth
ea
H

Remarks
of
es
pl

1. “Ability” is not choice variable.

ci
in
Pr

2. Occurrence of wrong diagnose no proof of lack

of ability.

3. Wrong diagnose is observable.

27-c
Three attitudes towards risk. Two alternatives: par-
ticipate in a risky situation → E(U ); do not partici-
pate → U (E(Y )).

lt
ira
G
z-
Def.: Risk aversion: E(U ) < U (E(Y )).

ne
ti
ar
M
Def.: Risk neutrality: E(U ) = U (E(Y )).

er
vi
Xa

Def.: Risk preference: E(U ) > U (E(Y )).

c
s
ic
om

U(Y)
on

Upr(Y) Unt(Y)
Ec

U(Yh) Uav(Y)
lth
ea

Uav[E(Y)]
H
of
es
pl

E(U)
ci

Unt[E(Y)]
in
Pr

Upr[E(Y)]
U(Ys)

Y
Ys E(Y) Yh

27-d
Example 1: tossing a coin
√
Individual: Y = 49 e, U (Y ) = Y.

lt
ira
Alternative 1. Participate in a lottery: toss a coin.

G
z-
ne
If win → 98 e. If loss → 0 e.

ti
ar
M
er
Cost of participation: 49 e.
vi
Xa

c

Expected utility:
s
ic
om

1 1
on

E(U ) = U (49 + 98 − 49) + U (49 − 49) =

2 2
Ec

1 1
lth

U (98) + U (0) ≈ 4.9497

ea

2 2
H
of
es

Alternative 2. Do not participate → U (49) = 7 (In

pl

this case, Y = E(Y ))

ci
in
Pr

Conclusion: E(U ) < U (E(Y ). Risk averse indi-

vidual decides not to participate.

Remark: U (Y ) is strictly concave.

27-e
How can the individual be induced to participate?

- 4 payment if winning: e.g. 256 e

lt
ira
G
1 1

z-
E(U ) = U (49 + 256 − 49) + U (49 − 49) =

ne
2 2

ti
1 1

ar
U (256) + U (0) =

M
2 2

er
1
16 = 8 > 7 vi
Xa
2

c
s
ic

-5 participation cost: e.g. 24 e

om
on

1 1
Ec

E(U ) = U (49 + 98 − 24) + U (49 − 24) =

2 2
lth

1 1
ea

U (147) + U (25) ≈
H

2 2
of

1 1
es

12.1243 + 5 ≈
pl

2 2
ci

6.0622 + 2.5 ≈ 8.56232 > 7

in
Pr

Remark: either way implies rising the expected value

of the lottery. The seller of the lottery tickets would
make a loss for sure if selling many tickets!

27-f
Example 2: contracting insurance

Individual with assets valued 21000 e.

lt
ira
Probability of losing 6000 e= 1%

G
z-
ne
ti
ar
Probability distribution:

M
er

1% −→ 15000 e
vi
Xa
99% −→ 21000 e

c
s
ic

Insurance: alter probability distribution

om
on
Ec

Insurance contract:
lth

- indemnity = 6000 e
ea

- premium = 60 e
H
of
es
pl

New probability distribution:

ci
in
Pr


1% → 20940 e(= 21000 − 6000 + 6000 − 60)
99% → 20940 e(= 21000 − 60)

Equal wealth in both states of nature: Individual fully

insured against loss.
27-g
Healthcare insurance

Risk aversion → contract healthcare insurance.

FDemand of insurance

lt
ira
Recall:

G
z-
• Individual: income Y , Utility U (Y ) concave.

ne
• Two states: healthy, sick (prob. p) → Yh, Ys

ti
ar
• L loss of income if sick.

M
er
• Protection against loss L → insurance indemnity:
Z e when sick. Premium: αZ e. vi
Xa

c
s
ic

How much insurance to buy? (i.e. choose the value

om

of Z to max E(U ))
on
Ec

Ys(Z) = Y − L − αZ + Z = Y − L + (1 − α)Z
lth

Yh(Z) = Y − αZ
ea
H

Formally,
of
es

max E(U ) = pU (Ys) + (1 − p)U (Yh)

pl

Z
ci
in

Solution:
Pr

∂E(U ) ∂U  ∂Ys ∂U  ∂Yh

 
=p + (1 − p)
∂Z ∂Y Ys ∂Z

∂Y Yh ∂Z
∂U  ∂U 
 
= (1 − α)p − α(1 − p) =0
∂Y Ys ∂Y Yh
28
Interpretation
Concave utility → decreasing marginal utility.
♠ Each extra euro of coverage implies higher in-
come when sick. But expected marginal utility of

lt
ira
each extra euro of coverage diminishes. Formally,

G
∂U diminishes as Z increases (marginal
(1 − α)p ∂Y

z-
s

ne
benefit).

ti
ar
♠ Each extra euro of coverage implies higher cost

M
er
(less income) when healthy. Thus, marginal income
vi
Xa
∂U increases as Z
increases. Formally, α(1 − p) ∂Y

h
c

increases (marginal cost).

s
ic

♠ Combination of these two opposite effects deter-

om
on

mines optimal demand of insurance.

Ec
lth

∂U
ea

∂Y ∂U
(1− p) α
H

α given ∂Y
h
of
es

€
pl
ci
in
Pr

∂U
p (1− α )
∂Y
s

Z
Z*
€
28-a
Highest premium willing to pay?

lt
ira
G
z-
Def.: Certainty equivalent (CE). Level of income whose

ne
utility is equal to expected utility, U (CE) = E(U ).

ti
ar
M
er
Highest premium = E(Y ) − CE
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

28-b
FSupply of insurance

Z ∗ depends on α. In turn, α is a decision of in-

surance company. Such decision depends on the

lt
structure of insurance market.

ira
G
z-
Assumption: perfectly competitive market.

ne
ti
ar
M
Expected profit = premia - indemnity payments

er
vi
E(B) = (1 − p)αZ − p(1 − α)Z = Z(α − p)
Xa

c

2 set-ups:
s
ic
om

Set-up 1: determine α (”fair premium”) solution of

on

E(B) = 0. Thus,
Ec
lth

α
b =p
ea
H

Interpretation
of

The fair premium is equal to the probability of falling

es
pl

sick. If insurer sets lower premium will incur (ex-

ci
in

pected) losses. If insurer sets too a high premium

Pr

will obtain (expected) extra profits → new entrants

offering lower premia.

Set-up 2: determine α solving maxα E(B). Solu-

tion depends on the market structure.
28-c
FEquilibrium of the insurance market

lt
ira
- Demand: [marginal benefit = marginal cost]

G
z-
ne
∂U  ∂U 
 
p(1 − α)  = α(1 − p)

ti
ar
∂Y Ys ∂Y Yh

M
er
- Supply
vi
Xa
α=p

c
s

Therefore, p(1 − α) = α(1 − p), and market equi-

ic
om

librium is characterized by
on
Ec

∂U  ∂U 
 
 =  .
lth

∂Y Ys ∂Y Yh
ea
H

Equality only satisfied when Ys = Yh, i.e.

of
es

Y − L + (1 − α)Z = Y − αZ, or
pl
ci

Z ∗ = L.
in
Pr

The individual optimally fully insured against expected

loss.

28-d
Demand of insurance and healthcare demand elas-
ticity

lt
ira
 D1 and D2 demands for healthcare of 2 illnesses.

G
z-
 Illness 2 more severe → more demand, more in-

ne
elastic.

ti
ar
 Price of healthcare: P

M
er
vi
Xa
N 2 scenarios: no insurance; insurance with copay-

c

ment c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

Higher elasticity → insurance generates higher rel-

ative welfare gain.

28-e
9. Contract theory

Introduction

lt
ira
So far, market failure → mkt power, ∆ returns, pub-

G
z-
lic goods, externalities.

ne
ti
ar
M
New element of analysis: private information (asym-

er
metric, imperfect).
vi
Xa

c

What is a contract? Bilateral agreement: contract-

s
ic
om

ing party (principal) delegates in contracted party

on

(agent) decision making.

Ec
lth
ea

Elements of a contract:
H

♦ Principal: offers contract; verifiable variables

of
es

♦ Agent: if accepts, performs effort for Principal.

pl
ci
in
Pr

Perfect agent: physician as perfect agent for the pa-

tient takes decisions (diagnostic, treatment) “as if”
it would be the very patient taking decisions should
(s)he have the same information as the physician.
→ deontologic code, hypocratic oath.
29
 lt
ira
- If there is no conflict, the agent behaves as if (s)he

G
would be the principal rather than himself.

z-
ne
- If conflict of interest, problem for the principal: make

ti
ar
sure that the agent (physician) respects the interest

M
er
of the principal (patient).
vi
Xa
- Usual scenario: conflict of interest between princi-

c

pal and agent.

s
ic
om
on

Conflict of objectives:
Ec

♠ salary: income for agent, cost for principal

lth
ea

♠ effort: benefits principal, costly for agent

H
of
es

INFORMATION?
pl
ci

Complete (perfect), incomplete (imperfect), symmet-

in
Pr

ric (public), asymmetric (private)

29-a
Definitions

Perfect information: at each move, party knows his-

lt
ira
tory of decisions so far.

G
z-
ne
ti
Imperfect information: not perfect.

ar
M
er
vi
Complete information: every party knows all rele-
Xa

vant information about other party (parties know de-

c
s
ic

cision tree).
om
on
Ec

Incomplete information: ∃ party uncertain about other

lth

party’s behavior, i.e. there are random elements in

ea

the relationship.
H
of
es
pl

Symmetric information: all parties have exactly the

ci
in

same information;
Pr

Asymmetric information: One party has more infor-

mation than the other party.

29-b
 lt
ira
Illustration 1: complete information

G
z-
ne
2 Hospitals deciding whether buying or not, new

ti
ar
MRI device.

M
er
vi
Xa
Payoff: patient share.

c
s
ic
om
on

H1 H1
b
Ec

nb b nb
lth

H2 H2
ea
H

b nb b nb b nb b nb
of
es

1/2 3/4 1/4 1/2 1/2 3/4 1/4 1/2

pl

1/2 1/4 3/4 1/2 1/2 1/4 3/4 1/2

ci
in

Complete and perfect information Complete and imperfect information

Pr

29-c
Illustration 2: incomplete information

H1 may be aggressive (pr. p) or soft (pr. (1 − p)).

lt
Nature determines. H2 does not know attitude H1.

ira
G
z-
ne
N

ti
p 1-p

ar
M
H1 H1

er
b nb b nb

vi
Xa
H2 H2

c

b nb b nb b nb b nb
s
ic
om

0.7 0.8 0.6 0.5 1/2 3/4 1/4 1/2

0.3 0.2 0.4 0.5 1/2 1/4 3/4 1/2
on
Ec

Incomplete and perfect information

lth
ea

N
H

p 1-p
of

H1 H1
es

b b
pl

nb nb
ci
in

H2 H2
Pr

b nb b nb b nb b nb

0.7 0.8 0.6 0.5 1/2 3/4 1/4 1/2

0.3 0.2 0.4 0.5 1/2 1/4 3/4 1/2

Incomplete and imperfect information

29-d
Provision of incentives and objectives in the health-
care sector
Objetive: Maximum quality with minimum cost.

lt
ira
G
z-
PATIENTS

ne
ti
ar
M
Problem 1:

er
vi
Xa
Healthcare insurance → limited cost sensibility, ex-

c
s

cessive visits, only issue is quality.

ic
om

Incentives: cost-sharing (copayments). [Avoid eq-

on

uity issues!]
Ec
lth
ea

Problem 2:
H
of
es

Degree of precision in following treatment → hard

pl
ci
in

to correct.
Pr

US (1995) losses due to deviations in use of phar-

maceuticals: 80,000 - 100,000 mil $

[recall efficacy vs. effectiveness (p. 6f)]

29-e
PHYSICIANS

Problem 1:

Market of experts → ”trust goods”: diagnostic and

lt
ira
treatment.

G
z-
Treatment: difficult to assess its quality (subject to

ne
random elements, every patient is different, ...).

ti
ar
Valuation can only be made by provider.

M
er
vi
Incentives: Separation between diagnostic and treat-
Xa
ment.

c

Example: Japan (Macho-Stadler, I., 1999, “Reflex-

s
ic
om

iones sobre la provisión de incentivos en los servi-

on

cios sanitarios”, p.5)

Ec

Incentives: Patient access to second opinions →

lth

discipline via reputation BUT search of info is costly.

ea
H
of

Problem 2:
es
pl

“Supplier induced demand” (capacity of the physi-

ci
in

cian to manipulate his income or the importance of

Pr

a service)

Incentives: Patient: access to second opinions; In-

surer: ∇ prices/services does not work. Ej.: Os-
akidetza (Macho-Stadler, I., 1999, p.6)
29-f
 lt
ira
G
z-
ne
Definitions of types of goods under asymmetric in-

ti
ar
formation

M
er
vi
Xa
- search goods: their quality is apparent before pur-

c

chase.
s
ic
om
on

- experience goods: their quality is apparent after

Ec

consumption.
lth
ea
H

- trust goods: their quality is not always apparent

of
es

even after consumption.

pl
ci
in
Pr

29-g
 lt
ira
Problem 3:

G
z-
ne
2 payment systems: fee-for-service, or capitation.

ti
ar
M
er
fee-for-service [fixed payment + cost reimbursement]:
vi
Xa
provider does not participate in costs; does not have

c

control on costs; does not avoid difficult (costly) pa-

s
ic
om

tients.
on
Ec

capitation [fixed payment per patient]: cost contention;

lth
ea

avoids costly patients.

H
of
es

Incentives: combination: fixed payment + partial cost

pl
ci

reimbursement → allows for cost control maintain-

in
Pr

ing quality.

29-h
HOSPITALS

2 payment systems: prospectives or retrospective

budgets.

lt
ira
G
z-
retrospective budgets [ex-post reimbursement of costs]:

ne
ti
no cost control.

ar
M
er
vi
prospective budgets [ex-ante]: cost control; quality
Xa

tends to ↓, unforeseen events (epidemic episodes ,

c
s
ic

catastrophic situations,...)?
om
on
Ec

Incentives: Prospective budget + payment accord-

lth

ing to comparative performance.

ea

Define “normal” price to each hospital service based

H
of

on average. Exclude special treatments.

es
pl
ci
in

Conclusion:
Pr

No incentive mechanism adequate for all situations:

Often, incentive mechanisms raise (unwanted) sec-
ond order effects, BUT better than doing nothing.
29-i
Asymmetric info and conflict of objectives. Example

Hospital (principal) contracts manager (agent) to de-

fend interests of hospital.

lt
ira
Hospital cannot perfectly control manager’s decisions.

G
Contract cannot be based on manager’s behavior

z-
ne
(not verifialble).

ti
ar
Hospital does not have info on manager’s charac-

M
er
teristics.
vi
Manager can exploit his informative advantage to
Xa

c

his own benefit, instead of hospital’s.

s
ic
om

Aim: study relation between two individuals, where

on
Ec

one of them has an informative advantage over the

lth

other and their objectives are not aligned. =⇒

ea

Provision of incentives to reach objective.

H
of

If interests would coincide, info would be communi-

es

cated eliminating asymmetry.

pl
ci
in
Pr

3 topics:
moral hazard
adverse selection
signalling
29-j
Moral hazard

In a moral hazard situation both parties have the

same info at the moment of signing the contract,
BUT afterwards the agent receives private info. The

lt
ira
principal cannot observe (verify) the effort (action)

G
z-
exerted by the agent.

ne
ti
ar
M
er
vi
Xa

c
s

Source: Macho-Stadler et al. (1994, p. 21)

ic
om

Examples
on
Ec
lth

F labor contracts: publisher representative to sell

ea

books. Only verifiable element: # books sold. Effort

H

(# hours visiting clients) not verifiable by publisher

of
es

→ payment cannot be dependent on effort.

pl
ci
in

F hospital: manager contracted to control costs. If

Pr

fixed payment → insufficient effort.

F researchers: research center contracts researcher

in a project. → difficult to distinguish a thinker from a
dreamer. Fixed payment → little incentives to think.
29-k
Example 1. Hospital with retrospective budget →
little incentives for cost containment
Naı̈ve solution: prospective budgets.

lt
ira
G
Example 2. Fully insured driver → little incentive for

z-
ne
careful driving.

ti
ar
Naı̈ve solution: “bonus-malus” system

M
er
vi
Example 3. Fully insured physician → little incentive
Xa

c

to exert the (costly) efficient level of effort to obtain

s
ic

best diagnostic.
om

Naı̈ve solution: make physician responsible for di-

on
Ec

agnostic errors. Reputation (cf. TV series “House”)

lth
ea
H

Naı̈ve solutions because too much risk on the agent:

of

(i) hospital may have high costs because unexpected

es
pl

4 unfortunate case mix, epidemic episodes ... (and

ci
in

for lack of effort)

Pr

(ii) driver may be unlucky on one occasion along the

year.
(iii) physician may obtain a wrong diagnostic by ac-
cident.
29-l
Too much risk on the agent. What to do?

♠ Hospital with prospective budget. Too a high risk

lt
ira
(earthquakes, epidemic episodes, ...)

G
z-
ne
ti
ar
Solution 1: Observe average performance along time

M
and implement compensations according to devia-

er
tions from the average value. vi
Xa

c
s
ic

Solution 2: Induce competition among hospitals. In-

om

direct method to achieve the same objective.

on
Ec
lth

♠ Hospitals with 6= case mix (because of the envi-

ea
H

ronment where they perform activity)

of
es
pl

Solution:
ci
in

Step 1: define types of illnesses with similar cost/patient

Pr

(Diagnostic related groups)

Step 2: Use unit average cost for each DRG and
compensate the hospital per patient within each DRG.

29-m
Moral hazard and demand for health care

Health Insurance: price patient 6= price treatment

Health insurance → ∆ price of health services

lt
ira
So far: L exogenous, BUT

G
demand sensible to price, Z ∗?

z-
ne
ti
ar
Assume (1):

M
Prob p sick, demand for treatment

er
vi
Prob (1 − p) healthy. No demand for health ser-
Xa

vices.
c
s
ic
om

We already know:
on

contract full healthcare insurance (cost of treatment

Ec

P1Q1); premium: pP1Q1

lth
ea

Assume (2):
H
of

Demand decreasing in price

es

full healthcare insurance → free treatment ⇒

pl
ci

∆ demand for healthcare until the level P = 0, say

in

Q2 > Q1.
Pr

Consequence:
Real cost of treatment: P1Q2 > P1Q1 ⇒

Problem:
29-n
♠ If insurer maintains premium, expected revenue

lt
ira
pP1Q1, expected payment pP1Q2 → losses.

G
z-
ne
♠ If insurer increases premium to pP1Q2 individ-

ti
ar
ual cannot contract insurance. (May buy premium

M
er
> pP1Q1 as protection against risk)
vi
Xa

c

Premium: 2 components
s
ic
om

- protection against risk,

on

- supplementary resources to compensate moral haz-

Ec

ard.
lth
ea
H

As before, Z ∗ → MR=MC
of
es

now, individual costs = premium + moral hazard.

pl
ci
in
Pr

♠ Insurer’s mechanisms to provide incentives to in-

surees: deductibles and copayments.

29-o
Effect of a deductible

Assume insurance contains a deductible of D e (Cost

borne by insuree before insurer starts covering expenses).

lt
Individual compares the level of services obtained

ira
G
after paying the deductible (Q2) and without insur-

z-
ne
ance (Q1).

ti
ar
M
Example

er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci

F no insurance: (P1, Q1)

in
Pr

F p: prob “accident”
F D = P1Q1
F individual obtains Q2 paying D(= P1Q1)
F benefit: area under demand curve between Q2
and Q1 (green area)
29-p
Insurer ∆D to D0 = P1Q3. Will the individual buy
the insurance?

lt
ira
G
z-
ne
ti
ar
M
er
vi
Xa

c
s
ic
om

♣ ∆ payment if sick = P1(Q3 − Q1) (blue+yellow)

on

♣ benefit: area under demand curve between Q2

Ec

and Q1 (blue+green)
lth
ea

♣ Summary:
H

- expense increase: yellow

of
es

- benefit increase: green

pl
ci
in
Pr

Conclusion:
If green > yellow → contract insurance with de-
ductible D0. Otherwise,
Too high a deductible → eliminates incentives to
contract insurance
29-q
Effects of a copayment

Initial situation: no insurance (P1, Q1)

Contract insurance with copayment c ∈ (0, 1) i.e.

lt
ira
P2 = cP1 ⇒ Demand increases Q1 → Q2

G
z-
tine
ar
M
er
vi
Xa

c
s
ic
om
on
Ec
lth

Value of services = P1Q2

ea

∆ expenditure = P1(Q2 − Q1) (blue+yellow)

H

∆ benefit = area under demand curve between Q2

of
es

and Q1 (blue)
pl

Triangle yellow: welfare loss → Individual demands

ci
in

more insurance services than optimum.

Pr

Interpretation
Insurance → consumer “as if” ignorant real cost of
health services → distortion in resource allocation
between demand for insurance and other goods.
29-r
Copayment and market equilibrium

Initial situation: no insurance (P1, Q1) with demand

= supply

lt
ira
Contract insurance with copayment c ∈ (0, 1) ⇒

G
z-
Demand increases Q1 → Q2

ne
ti
New equilibrium: (P2, Q2).

ar
M
er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H
of
es
pl

∆ expenditure = P2Q2−P1Q1 (blue+green+yellow)

ci
in

Resource allocation distortion:

Pr

∆ benefit induced by copayment = green

∆ costs additional demand = blue
Deadweight loss from overproduction of insurance
services = yellow
29-s
Adverse selection

Adverse selection appears in situations where the

agent has private information before signing the con-

lt
tract. In this case the principal can verify the agent’s

ira
behavior. Principal knows there are several types of

G
z-
agents but cannot identify it at the moment of the

ne
ti
contract.

ar
M
er
vi
Xa

c
s
ic

Source: Macho-Stadler et al. (1994, p. 23)

om
on

Examples
Ec

F Insurance company may face a potential client

lth
ea

with high or low risk. Insurer can design a contract

H

for each type of insuree, but does not know ex-ante

of

which is the optimal one.

es
pl
ci
in

F Regulation of a public monopoly. Theory: price

Pr

to marginal cost and cover fixed cost with a trans-

fer. → monopoly knows better its cost function than
regulator. Regulator includes monopolist informa-
tional advantge in the design of the contract (trans-
fer, price).
30
 lt
ira
G
F Patient has better info on his real health status

z-
ne
than insurer → excessive costs, higher probability

ti
ar
of falling ill.

M
er
vi
Xa
F Hospital has better info on the case mix of pa-

c

tients than insurer and Health authority → inflation

s
ic
om

of costs and budgets.

on
Ec

F Physician, after the visit, has better info on the

lth
ea

patient’s real health status than

H

- patient → induced demand, prescription of branded

of
es

drugs (instead of generics)

pl
ci

- hospital → excessive referral to specialized care.

in
Pr

30-a
 lt
ira
G
z-
Asymmetric info may cause the market to perform

ne
ti
poorly, and even disappear.

ar
M
er
vi
Asymmetric info, key element in insurance and health
Xa

care markets.
c
s
ic
om

Illustration: Akerlof’s (1970) lemons market.

on
Ec
lth

Used-cars on sale with 6= qualities.

ea
H

Sellers know about qualities better than buyers.

of

Lemons principle: Good cars are driven out of the

es
pl

market by the lemons

ci
in
Pr

30-b
Simplified Akerlof’s example (FGS, 2004, ch. 9):

lt
ira
• 9 used-cars qualities

G
z-
1 , 1 , 3 , 1, 1 1 , 1 1 , 1 3 , 2}
q = {0, 4

ne
2 4 4 2 4

ti
ar
M
• Uniform prob. of picking a car (= 1
9)

er
vi
Xa

• Sellers know quality

c
s
ic
om

• Buyers only know distr. qualities

on
Ec
lth

• (min) Reservation value to sellers= 1000q e

ea
H
of

• (max) Reservation value to buyers= 1500q e

es
pl
ci
in

• Auctioneer calls out market prices

Pr

• Sale at a price s.t. D = S

30-c
 lt
ira
G
Let p = 2000e per car

z-
ne
ti
ar
• Supply:

M
er
Sellers willing to sell each car if for each car,
vi
Xa
pb ≥ 1000 × q.

c
s
ic
om

car (q=2)→ 2000 = 1000 × 2 → S=9.

on
Ec

• Demand:
lth
ea

Average quality=1;
H

Buyers willing to buy if p ≤ 1500 × 1 = 1500;

of
es
pl
ci

(price)2000 > 1500(res.value) → D = 0.

in
Pr

30-d
Let p = 1500e per car

• Supply:
Sellers willing to sell each car if for each car,

lt
pb ≥ 1000 × q.

ira
G
z-
ne
pbq=2 = (1000)2 = 2000 > 1500 → not offered

ti
pbq=1.75 = (1000)1.75 = 1750 > 1500 → not

ar
M
offered

er
vi
pbq=1.5 = (1000)1.5 = 1500 = 1500 → offered
Xa
pbq=1.25 = (1000)1.25 = 1250 < 1500 → of-

c

fered
s
ic
om

Thus, → S=7 cars; average quality = 3/4.

on
Ec

• Demand:
lth

Average quality q̄ = 3/4;

ea

3 = 1125.
H

Buyers willing to buy if pq̄ ≤ 1500 × 4

of

Thus, (price)1500 > 1125(res.value) → D = 0.

es
pl
ci

etc, etc.
in
Pr

Conclusion:
Under asymmetric info, @p at which D = S.

Why? Lemons principle.

30-e
Assume symmetric info:

lt
ira
Buyers and sellers only know average quality (q̄ =

G
z-
1).

ne
ti
ar
M
Let p = 1500e per car

er
vi
Xa

• Supply:
c
s
ic

Sellers willing to sell if pbq̄ ≥ 1000q̄ = 1000;

om

(price)2000 > 1000(res.value) → S = 9.

on
Ec
lth

• Demand:
ea

Average quality q̄ = 1;
H
of

Buyers willing to buy if pq̄ ≤ 1500q̄ = 1500;

es
pl
ci
in

(price)1500 = 1500(res.value) → D = 9.
Pr

Equilibrium price of 1500 e and 9 cars are sold.

30-f
Akerlof and health insurance market.

Individual has more info on his real health status

than insurer.

lt
ira
G
If insurer ignores this fact, and set premium accord-

z-
ne
ing to general population statistics → losses. Why?

ti
ar
- high risk individuals more interested in contracting

M
er
insurance →
vi
Xa
- insurer’s customers will be a biased population sam-

c

ple.
s
ic
om

Insurers anticipates it → contracts with higher pre-

on
Ec

mia. Low risk individuals do not contract insurance:

lth

Exclusion
ea
H
of

Conclusion: asymmetric info → inefficient resource

es

allocation.
pl
ci

F ind. do not know their p → insurance against risk.

in
Pr

F If insurer offers same contract to everybody →

- low risk indiv, too high premium → underinsurance
- high risk indiv, too low premium → overinsurance.

Solutions
30-g
Solution 1: Screening

lt
ira
G
Insurer offers menu of contracts:

z-
ne
- i) contract with high coverage and high premium;

ti
ar
- ii) contract with low coverage and low premium

M
er
vi
Xa
Consequence: self-selection:

c

- low risk indiv, → contract ii)

s
ic
om

- high risk indiv, → contract i).

on
Ec

Problems
lth
ea

1. Argument OK if insurer is monopolist. 6 ∃ eq. if

H

competition.
of
es

2. Even when ∃ eq, it is inefficient: low risk indiv.

pl
ci

overinsured.
in
Pr

Solution 2: Signaling

30-h
Signaling

lt
ira
G
Similar situation to adverse selection. After knowing

z-
ne
his type and before signing the contract, the agent

ti
ar
may send a signal observable by the principal.

M
er
vi
Xa

c
s
ic
om
on

Source: Macho-Stadler et al. (1994, p. 24)

Ec
lth
ea

Example
H
of
es

F Physician shows credentials (Ph.D., etc) as sig-

pl
ci

nal of ability when being contracted by hospital/patient.

in
Pr

Also hangs from the walls of his office credentials so

that patients can see them.

31
Alternatively, the principal may posses private info

lt
that transmits to the agent through the contract de-

ira
G
sign.

z-
ne
ti
ar
M
er
vi
Xa

c
s

Source: Macho-Stadler et al. (1994, p. 25)

ic
om
on

Example
Ec
lth
ea

F University Dpts in job market include “goodies” in

H
of

offers as signals of quality.

es
pl
ci
in

What is a signal? investment to disclose some info

Pr

(the “type”) yielding some advantage over keeping it

secret.

31-a
Solution 2: Signaling

lt
ira
Low risk indiv. willing to show to insurer they are low

G
risk:

z-
ne
e.g. volunteer medical reports.

ti
ar
⇒ Signaling theory

M
er
vi
Xa
Problem

c

high risk indiv. want to look like low risk imitating

s
ic
om

their signals.
on
Ec

Consequence
lth
ea

- Insurers very cautions in interpreting signals

H

- As signaling is costly, low risk indiv. may prefer not

of
es

to signal.
pl
ci
in
Pr

Equilibria: 2 types
(i) Separating equilibrium
(ii) Pooling equilibrium

31-b
(i) Separating equilibrium

Appears when signaling is very costly for high risk

indiv. →
- high risk indiv. do not imitate

lt
ira
- Insurer takes signals serously

G
z-
- low risk indiv. obtain better contracts

ne
- signaling attractive for low risk indiv.

ti
ar
M
(ii) Pooling equilibrium

er
vi
Xa
Appears when imitation is not very costly →

c

- Insurer ignores signals

s
ic
om

- Signals are useless

on

- Nobody signals
Ec
lth

Problem (technical, but important)

ea
H

Even with high signaling costs, often also exist pool-

of
es

ing equilibria.
pl
ci
in

Examples:
Pr

- Corporation places ad looking for “young graduate”

→ education as a signal of ability.
- Hospital real “case mix” difficult to know by Health
authority.
- Physicians’ ability hard to know by patient.
31-c
Summary

Asymmetric Info gives rise to the possibility of ask-

ing the following questions:

lt
ira
G
z-
 A situation with relatively ignorant consumers pre-

ne
vents high levels of competition?

ti
ar
M
 Should be observe high price dispersion in the

er
health care market?
vi
Xa
 Will the health care market provide unnecessary

c

or non optimal treatments to patients?

s
ic
om
on

Objetive: find mechanisms to correct for the bad

Ec

consequences of asymmetric info.

lth
ea

Incentive contracts, competition, ...

H
of
es

BUT
pl
ci
in
Pr

Contracts require verifiable info → costly (monitor-

ing, auditing)

Competition may lead to exclusion of some groups

of individuals → equity, social justice
31-d
 lt
ira
G
z-
ne
ti
ar
M
er
Reading
vi
Xa

c

The Henry J, Kaiser Family Foundation, 2006, Illus-

s
ic

trating the Potential Impacts of Adverse Selection on

om
on

Health Insurance Costs in Consumer Choice Mod-

Ec

els, Snapshots: Health Care Costs,

lth
ea

http://www.kff.org/insurance/snapshot/chcm111006oth2.cfm
H
of
es
pl
ci
in
Pr

31-e
Supplier induced demand

WHAT IS IT?

lt
ira
Agency problem. Patient’s dependency on physi-

G
z-
cian gives physician an advantage (due to better

ne
info) to manipulate demand to his benefit.

ti
ar
M
er
Old topic (1958 →) Roemer’s effect: “a bed built is a
vi
Xa
bed filled” (Roemer 1961) Very high correlation be-

c

tween availability of beds/1000 inhabitants and oc-

s
ic
om

cupation rate (days in hospital/1000 inhabitants).

on
Ec

Generalization to physician services → SID

lth
ea

Well documented phenomenon, although controver-

H

sial
of
es

- introduction of wages controls on physicians, often

pl
ci

∆ SID
in
Pr

- clinical decisions often influenced by financial in-

centives (e.g. physicians with salaries, lower hos-
pitalization rates then physicians payed on fee-for-
services basis).

32
Basic model of SID

Initial situation (P1, Q1).

Supply increase S1 → S2

lt
ira
G
z-
ne
ti
ar
M
er
vi
Xa

c
s
ic
om
on

Standard micro analysis: eq: (P2, Q2). Total ex-

Ec

penditure 4 o 5 according to ε.
lth

Healthcare services: ε < 1 → Total expenditure ↓

ea

Moreover, 4] physicians → lower demand/physician.

H
of
es

SID: physicians use their capacity to induce extra

pl
ci

demand (D2). New eq. (P3, Q3).

in

If P3 > P1 SID with certainty (Reinhardt induce-

Pr

ment test).
If P3 < P1 ambiguity.

Modeling problem: analysis assumes competitive

market.
32-a
Other models of SID

1. Evans (1974) model (simplified)

Physician’s utility: U (Y, D)

lt
ira
Y = income; D = inducement (] h. induced demand)

G
z-
ne
ti
ar
M
er
vi
Xa

c
s
ic
om
on
Ec
lth
ea
H

Assume: π average profit rate

of

- max profit without inducement: πQ0

es
pl

- profit from inducement: πD

ci
in

Therefore: Y = πQ0+πD → combinations (Y, D)

Pr

feasible for physician.

Given U , eq. in A = (YA, DA)
? If 4 competition: π → π 0, π 0 < π →
New eq. with 4 inducement: DA → DB
0 0
aiming at π 0Q0 + π 0D ≥ πQ0 + πD.
32-b
2. SID vs. advertising

Begin: max. profit → M R1 = M C1 → (P1, Q1)

lt
P

ira
G
z-
MC2

ne
P2

ti
MC1

ar
M
P1

er
vi
Xa

c
s

D2
ic
om

D1
Q
on

Q1 Q2
Ec

MR1 MR2
lth
ea
H

Margin (P − CM ) incentive to stimulate demand

of

as each additional unit yields profits: D1 → D2 ∼

es

argument to introduce advertising.

pl
ci
in
Pr

BUT

inducing demand has cost (reputation, time): M C1 →

M C2
New equilibrium: (P2, Q2)
32-c
Remarks
1. Incentives to induce demand depends on the dif-
ference between P and M C (market power)
- if competitive market, small gap, low incentives;

lt
ira
- if high market power, big gap, high incentives

G
z-
- empirical evidence: physician, high monopoly power.

ne
ti
ar
M
Enhance competition to limit incentive to induce de-

er
mand.
vi
Xa

c

2. Professional and ethical considerations limit physi-

s
ic
om

cians activities (contracts between principal (patient)

on

and agent (physician)) → limits informational advan-

Ec

tage:
lth
ea

- monitor insurer activity

H

- ease possibility of second opinions

of
es

- improve patient’s info on treatments (internet)

pl
ci
in
Pr

3. Distinction between informative advertising (pos-

itive) and persuasive advertising (negative) corre-
sponds to a physician transmitting info to the patient
(positive) or implementing non optimal/unnecessary
treatments (negative).
32-d
10. Economic Evaluation

Optimal policy design requires evaluation to select

lt
ira
best alternative:

G
z-
ne
ti
ar
• What is included in the estimation of costs: the

M
costs at the hospital, the social costs, cost to the

er
individual? vi
Xa

c
s
ic

• We assign value to commodities through our choices

om

on the (free) “market-place”. But most markets are

on
Ec

neither free, nor ideal. To consent in a market is not

lth

identical to consent to a market. Can these consid-

ea

erations be included in the analysis? How?

H
of
es
pl

• Can we put a price on everything? is everything a

ci
in

commodity that is for sale if only the price is right?

Pr

life, honor, rights, embryos, human tissues...if not,

there are certain things that will be difficult or im-
possible to include in the economic analysis.

33
 lt
ira
G
General difficulty: assign value to human life

z-
ne
ti
ar
Quality-Adjusted Life Years (QALYs)

M
er
vi
Xa
Most often used measure of health services.

c
s
ic
om

- Consider an individual with health status Qt in year

on

t and life horizon T .

Ec

- Ideal health state is Q∗ (reference value)

lth
ea

- Let H(Q) be utility of individual with Q

H

- (Q1, t1; Q2, t2; . . . ; Qn, tn) health profile. Qτ pre-

of
es

vails tτ years, n τ =1 tτ = T
P
pl
ci

- r discount rate (interest rate)

in
Pr

33-a
Definition: QALY

lt
ira
G
z-
QALY index is a measure of the utility of health sta-

ne
ti
tus, U (Q, T ), defined as the present value of the

ar
M
utility of each health status weighted by the time pe-

er
riod along which that health status prevails:
vi
Xa
n

c

X Qτ tτ
U (Q1, t1; Q2, t2; . . . ; Qn, tn) =
s

τ
ic

τ =1 (1 + r)
om
on

Weight =1 if perfect health status. Weight = 0 if

Ec

death (negative as well)

lth
ea
H

Example 1
of
es

Indiv. 70 years old, life expectancy 20 more years.

pl

First 10 years, perfect health (weight = 1),

ci
in
Pr

Second 10 years, half quality of life.

QALY = (10 × 1) + (10 × 0.5) = 15.

33-b
Example 2: Esclerosis múltiple
Niveles de incapacidad:

lt
ira
1. Ausencia de incapacidad.

G
2. Ligera incapacidad social.

z-
ne
3. Incapacidad social grave y/o ligero deterioro del rendimiento

ti
ar
laboral. Capaz de realizar todas las tareas domésticas ex-

M
er
cepto las muy pesadas.
vi
Xa
4. Limitación muy grave en las posibilidades de elección de

c

trabajo y en el rendimiento laboral. Las amas de casa y los

s
ic
om

ancianos son tan solo capaces de realizar tareas domésticas

on

ligeras, pero son capaces de ir de compras.

Ec

5. Incapacidad para conseguir un empleo remunerado. Inca-

lth
ea

pacidad para proseguir cualquier tipo de educación. Ancianos

H

confinados en su hogar, excepto raras salidas acompañadas y

of
es

breves paseos, e incapaces de ir de compras. Amas de casa

pl

capaces sólo de realizar unas pocas tareas sencillas.

ci
in
Pr

6. Confinado en una silla de ruedas, o bien incapaz de des-

plazarse por la vivienda sin la ayuda de otra persona.
7. Confinado en cama.
8. Inconsciente.

33-c
 lt
ira
G
z-
Incapacidad Niveles de sufrimiento

ne
Ausente Leve Moderado Grave

ti
ar
Nivel 1. 1.000 0.995 0.990 0.967

M
Nivel 2. 0.990 0.986 0.973 0.932

er
Nivel 3. 0.980 0.972 vi 0.956 0.912
Xa
Nivel 4. 0.964 0.956 0.942 0.870

c
s

Nivel 5. 0.946 0.935 0.900 0.700

ic
om

Nivel 6. 0.875 0.845 0.680 0.000

on

Nivel 7. 0.677 0.564 0.000 -1.486

Ec

Nivel 8. -1.028 na na na
lth
ea
H
of
es
pl

Véase Hidalgo, A. (2000): Evaluación económica de tecnologı́as

ci
in

sanitarias, en Hidalgo et al. (2000, cap. 12).

Pr

33-d
Historia de un paciente:
- Sano durante 28 años
- Primeros sı́ntomas: 29 años
- Empeoramiento progresivo

lt
ira
- Muerte: 49 años.

G
z-
ne
ti
ar
M
Edad Incapacidad Sufrimiento Indice Valor

er
0-28 Sano Ausente 1.000 28.000
vi
Xa
29 nivel 2 Leve 0.986 0.986

c

30-33 nivel 2 Moderado 0.973 3.892

s
ic

34-36 nivel 3 Moderado 0.956 2.868

om

37-40 nivel 4 Moderado 0.942 3.768

on

41-43 nivel 5 Grave 0.700 2.100

Ec

44-46 nivel 6 Grave 0.000 0.000

lth

47-49 nivel 7 Grave -1.486 -4.458

ea
H

QALY 37.156
of
es
pl
ci

QALY = (28 ∗ 1) + (1 ∗ 0.986) + (4 ∗ 0.973)+

in
Pr

+ (3 ∗ 0.956) + (4 ∗ 0.942) + (3 ∗ 0.7)+

+ (3 ∗ 0) + (3 ∗ −1.486) = 37.156.
QALYs < 49 años: calidad de vida desde la aparición
de los sı́ntomas va en continuo retroceso.
33-e
Example 3. Eval. alternative health programs
Patient with health status 60% optimal.
If medication → 3 more years alive.
If surgery → 5 more years and better quality of life.

lt
Success rate of surgery: 40%.

ira
G
Prob. dying in surgery room: 3%

z-
Cost of surgery = 30, 000 e

ne
ti
Discount rate = 5%

ar
M
Year 1 2 3 4 5 total

er
Discount 1.00 0.95 0.91 0.86 0.82
vi
Xa
Medication

c

Life quality .60 .50 .40 .00 .00

s
ic

Disc. value .60 .48 .36 .00 .00 1.44

om

Surgery
on

Life quality .90 .80 .70 .60 .50

Ec

Disc. value .90 .76 .63 .52 .41 3.23

lth
ea

QALYs: Medic.= 1.44 QALYs; surgery = 3.23 QALYs

H

Risk of death w/ surgery = 3.23 × 0.03 = 0.09

of

Expected net ∆ QALY = [(3.23 − 1.44) × 0.4] −

es
pl

0.09 = 0.72 − 0.09 = 0.63 QALY

ci
in

i.e. surgery → expected net ∆ QALY = 0.63

Pr

Cost per extra QALY = 30, 000/0.63 = 47, 620 e

Conclusión: Patient if surgery, expected ∆ QALY of

0.63 years, at a cost of 47,620 e per QALY.
33-f
Ejemplo 3. Resource allocation
Society: 2 individuals, A, B
Initial situation: A1 = B1, (M )
Society gets extra resources → ∆ health → R =
max(A∗ + B ∗) s.t. (A∗, B ∗) ∈ health frontier

lt
ira
G
z-
B’s QALYs

ne
ti
ar
Bmax

M
B*

er
R

vi
N Xa
Q U

c
s
ic

Health frontier
B1
om

M
on
Ec

45º
0 Amax
lth

A1 A* A’s QALYs
ea
H

From R and along the frontier, 4 A’s QALY → 5

of

B’s QALY s.t. A + B ↓

es
pl

Is R socially efficient and/or egalitarian?

ci
in

As Bmax > Amax, the same resources will yield

Pr

more QALYs to B than to A.

Given that R = max(A∗ + B ∗), R is socially effi-
cient but, not egalitarian (B ∗ > A∗).
Egalitarian allocation: Q
Allocation max social welfare: N (not egalitarian)
33-g
Economic Evaluation
Questions for a planner:
♠ Who should do what to whom?
♠ With what health care resources?

lt
♠ With what relation to other health services?

ira
G
z-
Replies (Definition: Economic evaluation)

ne
ti
Comparative analysis of the use of resources (costs)

ar
M
and improvements from alternat. health programs

er
(consequences).
vi
Xa

c

Issues to analyze,
s
ic

Efficacy: Can the program work?

om
on

Effectiveness: Does it work?

Ec

Data: medical literature; but consider

lth
ea

- should adjustments be made?

H

- quality? Question checklist (McMaster Univ, ’81)

of
es

- relevance?
pl
ci
in

Availability: Does it reach the target population ?

Pr

Opportunity cost: is it worth compared with alterna-

tive uses of the same resources?

Optimality: Best possible use of resources?

34
Components in Economic Evaluation

lt
ira
G
z-
ne
ti
ar
M
er
vi
Xa

c
s
ic
om
on
Ec
lth

Consequences: 3 categories
ea
H

- Identification: Health effects (e) (4 QALYs)

of

- Valuation: utility (u), monetary (w), “info” (v)

es
pl

- Measurement (s): resources saved (s1, s2, s3)

ci
in
Pr

Costs
- health care sector (c1): hosp, physicians, pharma
- patient/family (c2): income, time
- other sectors (c3): opportunity costs
35
Economic Evaluation Techniques
Combination of costs and consequences

COMPARISON OF COSTS AND CONSEQUENCES

YES

lt
NO

ira
ONLY COSTS ONLY

G
CONSEQUENCES

z-
A

ne
L N COST-OUTCOME
O COST OUTCOME

ti
T

ar
DESCRIPTION DESCRIPTION DESCRIPTION
E

M
R

er
N

vi
A Xa
T
I

c

V COST-MINIMIZATION
EFFICACY
s

E Y
ic

S COST OR COST-EFECTIVENESS
E
om

S ANALYSIS EFFECTIVENESS COST-UTILITY

on

EVALUATION COST-BENEFIT
Ec
lth
ea
H

Cost minimization analysis

of
es

(c1 + c2 + c3) − (s1 + s2 + s3)

pl

Cost-effectiveness analysis (CEA)

ci
in

[(c1 + c2 + c3) − (s1 + s2 + s3)]/e

Pr

Cost-utility analysis (CUA)

[(c1 + c2 + c3) − (s1 + s2 + s3)]/u
Cost-benefit analysis (CBA)
(w + v + s1 + s2 + s3) − (c1 + c2 + c3)

36
Measurement of costs consequences

lt
ira
G
z-
ne
ti
ar
MEASUREMENT OF COSTS AND CONSEQUENCES

M
er
TYPE OF MEASUREMENT / IDENTIFICATION OF MEASUREMENT /
vi
Xa
STUDY VAL. OF COSTS CONSEQUENCES VAL. OF CONSEQUENCES

c
s

Min Cost $ IDENTICAL NONE

ic
om

SINGLE EFFECT OF
on

INTEREST COMMON NATURAL UNITS /

Ec

CEA $ TO ALL THE NONE

lth

ALTERNATIVES
ea
H

QALYs /
$
of

CUA SINGLE OR COST per QUALY

es

MULTIPLE EFFECTS
NOT NECESSARILY
pl
ci

COMMON TO ALL THE

in

CBA $ ALTERNATIVES $
Pr

37
 lt
ira
G
Cost minimization analysis

z-
ne
ti
ar
Illustration 1

M
er
vi
Xa
2 programs: minor surgery for adults

c
s
ic
om

Both accomplish successful surgery

on
Ec

Differ in effectiveness:
lth
ea

A: one night stay in hospital

H

B: day-surgery program
of
es
pl
ci

Evaluation: select the least cost alternative.

in
Pr

38
Cost minimization analysis (2)

Analysis: Four basic questions:

lt
ira
G
z-
ne
1. What costs should be considered?

ti
ar
M
♠ Health care sector

er
vi
♦ Hospital resources (treatment, bed days, out-patient
Xa

c

attendance, overheads)
s
ic

♦ Community care resources (GP visits, nurse vis-

om

its, ambulance)
on
Ec
lth

♠ Patient and family

ea

♦ Patients time: treatment and off-work

H
of

♦ Time of relatives looking after the patient

es
pl

♦ Out-of pocket expenses

ci
in
Pr

♠ Other sectors
♦ social workers visits, nursing home help, volun-
tary sector

38-a
2. What costs should be included?

♠ Viewpoint of analysis.
health sector?, hospital’s?, patient’s?, social?
e.g. monetary compensation while off-work:

lt
ira
♦ socially: no cost, no benefit (transfer)

G
z-
♦ Government: cost

ne
♦ Patient: revenue.

ti
ar
M
♠ Short/long run evaluation.

er
vi
♦ Intertemporal prefs (life is short, future uncertain)
Xa

♦ Illustration
c
s
ic
om

Year Cost A Cost B

on

1 5 15
Ec

2 10 10
lth

3 15 4
ea

Total 30 29
H
of
es

♠ Justify exclusion of “irrelevant costs”.

pl
ci
in

♠ Order of magnitude.
Pr

♠ Opportunity costs.

♠ No-market goods and services. (volunteer time,

donations)
38-b
3. How costs should be estimated?

lt
ira
♠ Is there a market?

G
YES: Opportunity costs (if there is info); Quantities

z-
ne
and prices

ti
ar
NO: negotiations; market values as proxy; shadow

M
er
prices. (Ethical values!)
vi
Xa

c

♠ Time span of project

s
ic
om

♦ costs ↓ in hospital but out-patient treatment needed

on

♦ costs ↑ in hospital but final (see previous exam-

Ec

ple)
lth
ea
H

♠ Assets
of
es

Equipment, buildings, land...

pl
ci
in
Pr

♠ Average vs Marginal costs

Compute present values

38-c
Discounting future costs to present values

lt
ira
(recall example page 35b)

G
z-
N

ne
Fn(1 + r)−n
X
P =

ti
ar
n=1

M
er
where,
vi
Xa
P : present value

c

Fn: future cost at year n

s
ic

r : annual interest (discount) rate

om
on

Take r = 5%.P A = 26.79 < P B = 26.81 !!

Ec

Note: expenses computed at the end of period.

lth
ea
H

Choice of r: praxis {0, 3%, 5%}

of
es

social opportunity cost

pl

social rate of time preference

ci
in
Pr

convention: 5%
shadow-price of capital:3%

38-d
 lt
ira
G
z-
ne
ti
ar
M
4. How accurate does costing should be?

er
vi
Xa

♠ Availability of data
c
s
ic
om

♠ Time and effort vs. outcome

on
Ec
lth

“Do not make the perfect the enemy of the merely

ea
H

good”
of
es
pl
ci
in
Pr

38-e
Shadow prices

Concept linked to optimization problem:


g(x, y) < k

lt
max f (x, y) s.a

ira
x,y h(x, y) < q

G
z-
ne
max L(x, y) = f (x, y)+λ(k−g(x, y))+µ(q−h(x−y))

ti
x,y

ar
M
er
Illustration
vi
Xa

c

f (x, y): number of patients treated

s
ic

x: available equipment
om

y: labor
on
Ec

k: budget constraint
lth

q: available time
ea
H

y
of
es

h(x,y)
pl
ci
in
Pr

f(x,y)=m

Feasible g(x,y)
set

38-f
 lt
ira
G
λ and µ represent rate of change of f (x, y) wrt

z-
ne
change in k and q;

ti
ar
λ and µ represent additional resources needed so

M
er
that objective function achieves value beyond m;
vi
Xa
λ and µ evaluate that cost;

c

λ and µ called shadow prices

s
ic
om
on

λ > 0 ; constraint g(x, y) < k binding.

Ec

µ > 0 : constraint h(x, y) < q binding.

lth
ea
H

λ = 0 ; constraint g(x, y) ≤ k not binding.

of
es

µ = 0 : constraint h(x, y) ≤ q not binding →

pl
ci

resource requirement ≤ availability

in
Pr

38-g
Example

Recall example in pp. 5i-5j.

lt
First order conditions,

ira
G
∂L

z-
= αxα−1y β − λPx = 0 (8)

ne
∂x

ti
ar
∂L
= βy β−1xα − λPy = 0

M
∂y

er
∂L vi
Xa
= m − xPx + yPy = 0

∂λ
c
s

Substituting values of
ic
om

αm
on

x(Px, m) =
Px(α + β)
Ec

βm
lth

y(Py , m) =
ea

Py (α + β)
H
of

in, say, (8) we obtain

es

α−1 β
pl


αm βm
ci

αxα−1y β α P (α+β) Py (α+β)

in

x
λ= =
Pr

Px Px
If α = 1/3, β = 2/3, m = 30, Px = 2, Py = 1,
 2/3
20
3
λ= ≈ 0.202
6(5)2/3
38-h
Cost-Effectiveness Analysis

lt
ira
G
z-
Illustration

ne
ti
ar
M
♣ 2 programs:

er
- A: hospital dialysis vi
Xa

- B: kidney transplantation
c
s
ic
om

♣ Outcome of interest (common):

on
Ec

life-years gained after renal failure.

lth
ea

♣ Programs Differ in success and in cost.

H
of
es
pl

♣ Comparison:
ci
in

- Cost per life-year gained

Pr

- Life-years gained per dollar spent

39
Cost-Effectiveness Analysis (2)

Basic elements
♠ Choosing a measure of effectiveness

lt
♦ define, first, objectives of the program

ira
G
♦ performance of program in actual use

z-
ne
♦ relate to

ti
 final health output (life-years gained)

ar
M
 intermediate output (cases found, patients appro-

er
vi
priately treated) if value by itself
Xa

c

♠ Linking intermediate and final outcomes

s
ic
om

♦ important in prevention studies

on

♦ confident in proper link (previous research)

Ec
lth

♠ Discounting of effects (as well as costs)

ea

♦ if only costs, may yield nonsense conclusions

H
of

♦ if only costs, present value of the cost next year

es

is always lower than today

pl
ci

♦ if only costs, differential treatment with respect to

in
Pr

other sectors in the economy may yield inconsisten-

cies in the overall allocation of resources
♦ individuals can trade reductions in health status
and other goods/services today in return for a health-
ier time in the future (and vice versa)
39-a
♠ Uncertainties and Sensibility Analysis
♦ no data may be available → “informed guess”
♦ estimates may not be precise

lt
ira
♦ methodological controversies

G
♦ Sensibility: 3 steps

z-
ne
 identify uncertain parameters for which sensitivity

ti
ar
analysis is required

M
er
 specify plausible range of these parameters
vi
Xa
 compute results based on best, optimistic and

c

pessimistic guesses
s
ic
om
on

Use of Quality-Life Scales

Ec
lth
ea

♠ Additional measure of effectiveness

H

♠Relative costs of treatments compared with their

of
es

relative consequences in natural units (success of

pl
ci

therapy, life-years gained)

in
Pr

♠ 2 types of quality-life scales

♦ specific (disease specific, age specific)
♦ general health profiles: Short Form 36, Notting-
ham Health Profile, Sickness Impact Profile

39-b
 lt
ira
General health profiles

G
z-
ne
 physical functioning

ti
ar
 ability for self-care

M
er
 psychological status
vi
Xa
 level of pain and distress

c

 social integration
s
ic
om
on

Advantages:
Ec

reliable and valid, supplementary information only.

lth
ea
H

Disadvantages:
of
es

- multidimensional
pl
ci

- not based on individuals preferences

in
Pr

- not possible to combine quality and quantity of life

39-c
Cost-Utility Analysis

lt
ira
G
z-
Illustration

ne
ti
ar
M
♠ Measure of value: utility of effects.

er
vi
Xa

♠ Twins identical except in occupation (sign painter,

c
s
ic

soccer player).
om
on
Ec

♠ Valuation in [0,1] of broken arm will differ

lth
ea
H

♠ Utility of treatment will differ

of
es
pl

♠ CUA allows for quality of life adjustments via QALYs:

ci
in

generic outcome measure for comparison of costs

Pr

and outcomes in different programs.

40
Cost-Utility Analysis (2)

Basic elements

lt
ira
G
♠ Form of evaluation that places particular attention

z-
ne
on the quality of the health outcome produced.

ti
ar
♠ CUA vs. CEA

M
♦ CEA: evaluation measured in natural units (blood

er
vi
pressure improvement, patients improved, lives saved,
Xa

c

life-years gained, etc.)

s
ic

♦ CUA: measured in QALYs gained.

om

♦ Identical on the cost side. Different on the out-

on
Ec

come side (see p.33)

lth

♦ CEA: its specificity makes comparisons across

ea

studies difficult.s
H
of

♦ CUA: generic measure of outcome allows for com-

es
pl

parisons. Highlights crucial role of preferences (util-

ci
in

ities) in valuing the outcomes.

Pr

♠ CUA is useful if:

♦ health-related quality of life is important,
♦ program affects both mortality and morbidity,
♦ program to be compared are multidimensional.
40-a
 lt
ira
G
z-
Utility function (von Neumann-Morgenstern)

ne
ti
ar
M
♠ Utilities ≡ Preferences:

er
vi
Xa

♦ Ordinal: ranking
c
s
ic

♦ Cardinal: numbers representing strength of pref-

om

erences for the outcome relative to the others.

on
Ec
lth

♠ Environment:
ea
H
of

♦ Certainty
es
pl

♦ Uncertainty: risk aversion, risk neutrality, risk love

ci
in
Pr

40-b
Measuring preferences:
ask subjects to rank health outcomes, i.e. construct
a rating scale.

♠ Interpretation: ‘The difference in desirability be-

lt
ira
tween outcomes A and B is twice as great as the

G
z-
difference between C and D. Hence, I will make the

ne
interval between A and B twice as large”.

ti
ar
M
♠ Example:

er
Standard gamble (classical method to measure car- vi
Xa

dinal preferences)
c
s
ic
om

healthy for t years (u1 )

p
on
Ec
lth ive 1

1−p
ea ernat

death (u2 )
H Alt

(u1 > u2 > u3 )

Al
ter
of
na
tiv
es

2 e

state i for t years with certainty (u3 )

pl
ci
in

Problem: find p such that the individual is indifferent

Pr

between both alternatives.

u2−u3 : cardinal utility index.
Probability p = u1−u3

♠ QALYs (see pp. 30a-30d)

40-c
Cost-Benefit Analysis

Example 1 (multiple common effects)

♣ 3 programs:

lt
ira
- A: hospital dialysis

G
z-
- B: kidney transplantation

ne
- C: home dialysis

ti
ar
M
er
♣ Outcomes of interest (common):
vi
Xa
- Life-years gained

c

- Quality of Life
s
ic
om

- Incidence of Medical Complications

on
Ec

♠ Compute cost effectiveness ratios

lth
ea
H

♠ NO alternative superior on 3 dimensions?

of
es

♦ Determine a primary effect and CEA, or

pl

♦ Attach a money measure of value to all effects

ci
in
Pr

resulting from each program

♠ Pro: directly comparable w/ costs

♠ Con: difficult to translate effects into e

41
 lt
ira
G
Example 2 (multiple not common effects)

z-
ne
ti
ar
♣ 2 programs:

M
er
vi
Xa
- A: hypertension screening

c
s
ic
om

- B: influenza inmunization
on
Ec

♣ Outcomes of interest:
lth
ea
H

- A: prevention of premature death

of
es
pl
ci

- B: prevention disability days

in
Pr

41-a
Cost-Benefit Analysis (2): Measuring benefits

lt
ira
2 alternatives

G
z-
ne
ti
♠ Human capital approach

ar
M
er
vi
♦ Use monetary value of lost productivity
Xa

c
s
ic

♦ Problem: how to measure benefits at retirement

om

age? It is not acceptable that there are no benefits!

on
Ec

(cynical view: pensions saved)

lth
ea

♠ Willingness to pay
H
of
es
pl

♦ Closer to the notion of opportunity cost

ci
in
Pr

♦ More difficult to measure (but this an implemen-

tation argument)

41-b
Basic open issues

 Are ethical, cultural values pre-determined? Or

lt
ira
G
do they interact?

z-
ne
ti
ar
 What is the meaning of “fair” and “just”?

M
No definite answer is available; it influences the way

er
we want to measure benefits vi
Xa

c
s
ic

 include distribution concerns?

om

Requires detailed information, presumably difficult

on
Ec

to obtain
lth

On conceptual grounds, offers no problem

ea
H

Example:
of
es

SW = w(y)B(y)dF (y)
pl
ci
in

where
Pr

- F (y): cumulative distribution of income

- B(y): benefits
- w(y): weight to each income level

41-c
 lt
ira
G
z-
ne
Distribution concerns

ti
ar
M
er
♣ Usually economic evaluation does not treat ex-
vi
Xa
plicitly the distribution of consequences and costs

c
s
ic
om

♣ Options made have implicit treatment:

on
Ec

- Willingness to pay (richer more willing to pay), gives

lth
ea

more weight to rich people

H
of
es

- QALYs: same weight to everyone

pl
ci
in
Pr

42
Additional remarks

lt
ira
G
z-
♣ Economic evaluation assumes that freed resources

ne
ti
are redeployed efficiently.

ar
M
er
vi
Xa

♣ Incentives: patients with kidney problems follow-

c
s
ic

ing their treatment correctly (incorrectly)

om
on
Ec

→ are the ones with least (most) transplantations.

lth
ea

⇒ Bad behavior yields more severe health status

H
of

and go first in the waiting list.

es
pl
ci
in

♣ BUT if we aim at maximize the aggregate life

Pr

years, may be OK!

43
11. Macroeconomics

What is Macroeconomics about?

lt
ira
Global behavior of the economy. Aggregation.

G
z-
ne
Macro variables: GDP, Aggregate expenditure, Un-

ti
ar
M
employment, Inflation, Consumption, Saving, Invest-

er
ment, Exports, Imports, Public expenditure, etc.
vi
Xa

c

Questions:
s
ic
om
on

Long term growth,

Ec
lth
ea

Economic cycles,
H
of
es

Unemployment,
pl
ci
in
Pr

Inflation,

International Trade and Development,

Economic Policy (monetary, fiscal, labor, etc).

44
Definition of GDP: market value of all final goods
and services produced within a country in a given
period of time. [“Gross” means depreciation of cap-
ital stock included]

lt
ira
G
z-
Measures of GDP: Two approaches: expenditure

ne
and income (equivalent)

ti
ar
M
er
GDP - expenditure: adding up expenditure on all
vi
Xa
final goods and services produced during the year.

c
s
ic

GDP = private consumption + investment

om
on

+ government spending + (exports - imports)

Ec

≡ C + G + I + (X − M )
lth
ea
H
of

GDP - income: adding up all payments to owners

es

of resources used to produce output during the year

pl
ci

(aggregate income)
in
Pr

GDP = W + P + (T − S)
Equivalence in the National Income Accounts,

W +P +(T −S) = GDP = C +G+I +(X −M )

45
 lt
ira
Private consumption (C): commodities and services

G
z-
acquired by households.

ne
ti
ar
M
Investment (I): goods and services increasing the

er
vi
capital stock. Investment = Savings.
Xa

c
s
ic

Public consumption (G): goods and services acquired

om

by the public administrations (army, roads). No trans-

on
Ec

fers (pensions, social programs) because these are

lth

transfers.
ea
H
of

Net exports (X-M): net spending from rest of the

es
pl

world in goods and services yielding income to na-

ci
in

tional producers.
Pr

45-a
 lt
ira
G
z-
ne
Wages and salaries (W): Compensation of employ-

ti
ar
ees measures the total remuneration to employees

M
for work done. It includes wages and salaries, as

er
vi
well as employer contributions to social security and
Xa

other such programs.

c
s
ic
om

Profits (P): Surplus due to owners of incorporated

on
Ec

businesses. Often called profits.

lth
ea
H

Net taxes (T-S): Difference between the resources

of

transferred from the families to the State and the

es
pl

transfers from the State to the families.

ci
in
Pr

45-b
 lt
ira
Illustration:

G
z-
ne
ti
real GDP components in Spain in 2006 (constant

ar
M
prices 2000)

er
Demand components vi
Xa

c
s

106 e %
ic
om

Private consumption (C) 553.867 56.7

on

Public consumption (G) 184.233 18.9

Ec

Investment (I) 298.362 30.6

lth

Exports 254.985 26.1

ea
H

Imports 315.258 -32.3

of

Net exports (X-M) -60.273

es

TOTAL 976.189 100

pl
ci
in

Source: Contabilidad Nacional de España, INE.

Pr

45-c
 lt
ira
G
Illustration (2):

z-
ne
ti
ar
real GDP components in Spain in 2006 (constant

M
er
prices 2000)
vi
Xa
Supply components

c
s
ic

106 e %
om

Agriculture 27.199 2.8

on
Ec

Industry 151.709 15.5

lth

Construction 106.437 10.9

ea

Services 583.773 59.8

H

Net taxes 107071 11.0

of
es

TOTAL 976.189 100

pl
ci

Source: Contabilidad Nacional de España, INE.

in
Pr

45-d
Circular flow model:
Flow of resources, products, income, and revenue
among economic decision makers.

lt
ira
G
z-
ne
Rest
)

ti
(X of the

ar
rts World
po

M
Ex
8

er
C+I

)
C+

(M
6
vi I +G
Xa
(I)

rts
+(
t

po
en
)
(C

X-

Im

stm

M
n
tio

)=
ve

GD
In
mp

7
ic

P
su

9
n

om
Co

Financial
)
(G

markets
on

s
ase

5 10
Ec

S)
Go

h
urc

(
gs
vt's

n
lth

vi
vt.
loa

Households Sa Firms
Go
ea

ns
H
of

4 Government 1
Ta
es

xes
s
fer 2
pl

s
Di

an
ci

Tr
sp

P
GD
in
os
ab

d=
Pr
le

an
inc

3
em
om

.D
gr
e

Ag

45-e
Legend:

Flows of income:

lt
ira
G
z-
ne
(1): GDP = Aggregate income.

ti
ar
(2): Taxes are transfers from families to the State.

M
(3): Transfers from the State to the families.

er
vi
(4): Disposable income of families = Aggregate income-
Xa

taxes+transfers.
c
s
ic
om

Flows of expenses:
on
Ec
lth

(5): Disposable income splits in consumption and

ea
H

savings (= investment).
of

(6): Investment adds to flow of expenditure.

es
pl

(7): Public expenditure adds to flow of expenditure.

ci
in

(8): Exports add to flow of expenditure from the rest

Pr

of the world.
(9): Imports are transfers to the rest of the world.
(10): National account identity.

45-f
Example

Orange Inc. Juice Inc.

Wages and salaries 15000 Wages and salaries 10000

Taxes 5000 Taxes 2000

lt
ira
Purchase of oranges 25000

G
z-
Revenues from oranges 35000 Revenues from juice 40000

ne
Consumers 10000
Juice Inc. 25000

ti
ar
M
Profits before taxes 20000 Profits before taxes 5000
Profits after taxes 15000 Profits after taxes 3000

er
vi
Xa

c

VA Orange Inc. = 35000 (revenues from oranges)

s
ic
om

VA Juice Inc. = 40000 - 25000 = 15000 (revenues

on

from juice - cost of oranges)

Ec
lth
ea

VA total= 35000 + 15000 = 50000 = GDP (produc-

H

tion)
of
es
pl

GDP (income): 10000 + 40000 = 50000 (consumers

ci
in

expenditure)
Pr

GDP (income): (15000 + 10000) + (20000 + 5000)=

50000 (wages + profits before taxes)

Total production = Total income = Total expenditure

45-g
real GDP and nominal GDP.

lt
ira
G
nominal GDP: market value of production at today’s

z-
ne
prices.

ti
ar
M
er
Example: Economy with two goods (apples and or-
vi
Xa
anges)

c
s
ic

2006 ∗Q2006 )+(P 2006 ∗Q2006 )

GDP n2006 = (Pora
om

ora app app

on
Ec

Problem: If prices double, GDP also doubles =⇒

lth
ea

poor welfare indicator.

H
of
es

real GDP: market value of production at prices of a

pl
ci

reference year (1996).

in
Pr

GDP r2006 = (Pora

1996 ∗Q2006 )+(P 1996 ∗Q2006 )
ora app app

45-h
 lt
ira
Illustration:

G
z-
ne
Evolution GDP n and GDP r Spain 1995-2003 (106

ti
ar
e).

M
er
vi
Xa
Year GDPn price index GDPr

c

1995 437.783 100 437.783

s
ic

1996 464.251 103.5 448.457

om

1997 494.140 105.9 466.513

on
Ec

1998 527.975 108.5 486.785

lth

1999 565.419 111.4 507.346

ea

2000 610.541 115.3 529.691

H

2001 653.927 120.1 544.496

of
es

2002 698.589 125.5 556.651

pl

2003 744.754 130.5 570.556

ci
in
Pr

Source: INE.

45-i
The working of the Economy

lt
ira
G
Model of aggregate demand and supply:

z-
ne
ti
ar
(1) understand incidence of the different forces on

M
er
macro variables, and
vi
Xa

c

(2) measure potential effectiveness of economic poli-

s
ic
om

cies.
on
Ec

Aggregate demand is (the value of) the total quan-

lth
ea

tity the different sectors of the economy are willing

H

to spend in a particular period.

of
es
pl
ci

Graphically, market demand curve: relation between

in
Pr

general price level of the economy and aggregate

spending in goods and services in the economy.

46
Aggregate supply: (value of the) total quantity of
goods and services firms in the country are willing
to produce in a given period.

lt
ira
The market supply curve shows the production level

G
z-
firms are willing to supply at any given price level.

ne
ti
ar
M
Macroeconomic equilibrium: characterization of the

er
production level and of the price level.
vi
Xa

c

Graphically: intersection point of aggregate demand

s
ic
om

and supply curves. Compatibility between consumers

on

and producers behavior.

Ec
lth
ea

Equilibrium: two (potential) problems

H

1. negative results: equilibrium price-production pair

of
es

may not satisfy macro objectives (inflation, unem-

pl
ci

ployment, investment level, ...)

in
Pr

2. unstable results: even if the economy reaches

optimal equilibrium may be perturbed by external
shocks. oil crisis, bird flu, ...

=⇒ MACROECONOMIC POLICY
46-a
Aggregate supply and demand curves are shifted by
changes in consumers and/or producers behavior
(endogenous and/or exogenous shocks).

lt
ira
Options of the macroeconomic policy:

G
z-
ne
ti
1. shift demand curve through fiscal and monetary

ar
M
policy;

er
vi
Xa

2. shift supply curve through R&D policies;

c
s
ic
om

3. do nothing if the causes of the perturbation are

on
Ec

not identified.
lth
ea
H
of

P Supply P
Supply
es

Supply
pl

shock
ci

Economic
in

policy
Pr

action

P* P* Demand
shock
Economic
Demand policy Demand
action
Q* Q Q* Q

46-b
Unemployment.

Objective: maximize employment level.

lt
ira
Active population: set of people legally able to work

G
z-
= employed + unemployed.

ne
ti
ar
M
Activity rate: (employed/active pop.)*100

er
vi
Xa
Unemployment rate: (unemployed/employed)*100.

c
s
ic
om

Labor market equilibrium: wage level inducing com-

on

patibility between labor supply and demand.

Ec
lth
ea

Full employment 6= absence of unemployment −→

H

frictional unemployment (3% − 5%).

of
es
pl
ci

Structural unemployment: lack of adjustment be-

in
Pr

tween labor demand and supply. (labor market rigidi-

ties, professional qualifications, ...)

Frictional + structural unemployment = involuntary

unemployment = unemployment rate.
47
Measuring Unemployment.

lt
ira
“Encuesta de Población Activa” (Active population

G
enquiry): estimated unemployment [harmonized across

z-
ne
OECD countries].

ti
ar
M
er
Sampling on population −→ number of employed,
vi
Xa
unemployed, discouraged, by age, sex, education

c

level, length of unemployment, etc.

s
ic
om
on

Def.: unemployed individual not working the previ-

Ec

ous week, but ready to take a job along the following

lth
ea

two weeks.
H
of
es

Def.: employed individual with a job (≥ 1 hour) in

pl
ci

the previous week.

in
Pr

Active population = population employed + popula-

tion unemployed.

47-a
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G
ira

47-b
lt
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G
ira

47-c
lt
Inflation.

Sustained and generalized increase of the general

level of prices of goods and services in an economy.

lt
ira
G
z-
How to define that price level? → Two alternative

ne
price indices (weighted average of prices):

ti
ar
M
er
1. GDP deflator,
vi
Xa

c

2. CPI (Consumer price index).

s
ic
om
on

? GDP deflator
Ec
lth
ea

GDP deflator = (nominal GDP)/(real GDP).

H
of
es

In our economy with oranges and apples,

pl
ci
in

2006 ∗ Q2006 ) + (P 2006 ∗ Q2006 )

(Papp app ora ora
Pr

GDP deflator = 1996 ∗ Q2006 ) + (P 1996 ∗ Q2006 )

.
(Papp app ora ora

Comparison of a consumption bundle evaluated at

today’s prices and at the base year prices.
48
? Consumer Price Index.

lt
ira
G
CPI = nominal value of consumption bundle/real value

z-
ne
of that consumption bundle.

ti
ar
M
er
consumption bundle: “Encuesta de Presupuestos
vi
Xa
Familiares del INE” → representative sample of con-

c

sumption goods of families weighted by their impor-

s
ic
om

tance.
on
Ec

CPI evolution: monthly, yearly, aggregated within

lth
ea

the year, last 12 months.

H
of
es

Disaggregate CPI in sectorial price indices: non-

pl
ci

energy goods and services, energy goods and ser-

in
Pr

vices, non-manipulated food, ... → analyze their

evolution. (See sample figures)

48-a
 Pr
in
ci
pl
es
of
H
ea
lth
Ec
on
om
ic
s
c

Xa
vi
er
M
ar
tine
z-
G
ira

48-b
lt
 lt
ira
G
z-
ne
ti
ar
M
er
vi
Xa
CPI and price index of non elaborated produced goods and services

c

excluding fats, tobacco and touristic packs.

s
ic
om
on
Ec
lth
ea
H
of
es
pl
ci
in
Pr

Price index of non elaborated produced goods and services excluding fats
and tobacco (BENE-X), and price index of services excluding touristic
packs (SERV-T).

48-c
CPI vs. GDP deflator

1. GDP deflator measures the prices of all goods

and services produced.

lt
ira
CPI measures prices of goods and services in the

G
z-
representative consumption bundle.

ne
ti
ar
M
2. GDP deflator considers only goods and services

er
produced inside the country. vi
Xa

c
s
ic

3. CPI is computed for a fixed consumption bundle;

om

GDP deflator allows for variations of the bundle along

on
Ec

time in accordance with the variation in the compo-

lth

sition of the GDP.

ea
H
of

4. CPI does not measure possibility of consumers

es
pl

to alter the composition of the bundle (neither sub-

ci
in

stitution nor income effects.

Pr

Although CPI may differ from GDP, both convey the

same info on the rhythm of price increase. See next
figure.
48-d
 lt
ira
G
Illustration:

z-
ne
ti
ar
%

M
er
CPI

vi
Xa

c
s
ic

GDP
om

deflator
on
Ec
lth
ea
H
of
es

Year
pl
ci
in

Source: US Department of Commerce, Department of Labor.

Pr

48-e
The Phillips curve.

Inverse relationship between inflation rate and un-

employment rate. Controversial!!!

lt
ira
Reductions of unemployment rate against increases

G
z-
in inflation rate;

tine
ar
If prices moderate their increment, will yield an in-

M
crease in unemployment.

er
vi
Xa

c
Inflation

s
ic
om
on
Ec

Phillips curve
lth
ea
H
of
es

unemployment
pl
ci
in

The Natural Unemployment Rate

Pr

In the long term the economy tends towards an un-

employment rate independent of the implementation
of fiscal and/or monetary policies (with only short
run effects).
48-f
Illustration: US 1960-1995.

lt
ira
Inflation rate

G
z-
ne
ti
ar
M
er
vi
Xa
Unemployment rate

c
s
ic

∗ 1960-1969: good fit (increasing inflation). Average

om

inflation 2.5%, unemployment 4.8%

on
Ec
lth

? 1970-1973: change in expectations (curve shifts).

ea

Average inflation and unemployment 5.2%

H
of
es

◦ 1974-1983: oil shock. Worse fit. Average inflation

pl
ci

8.2%, unemployment 7.5%

in
Pr

• 1984-1995: improve expectations. Average infla-

tion 3.7%, unemployment 6.2%

Controversy: curve shifts vs. existence of the curve.

48-g
Macroeconomics of the health care market

Variables: 4 groups

lt
ira
G
z-
? population health status

ne
ti
ar
- Life expectation at birth

M
- Mortality rate

er
vi
- other: quality of life, morbidity, ...
Xa

c
s
ic

? Life style and behavior

om

- consumption of tobacco, alcohol and other drugs

on
Ec

- other: education rate, ...

lth
ea

? Level of health services

H
of

- health expenditure per capita

es
pl

- other: % hospital and pharma expenditure, ...

ci
in
Pr

? Health promotion
- % health care over GDP
- other: number physicians, nurses, ...

49
Relation between macro and health variables

F Economic growth

lt
 Positive effects on health:

ira
G
z-
- Life expectancy at birth: Spain 1960-97. 4 in 8

ne
ti
years (70 to 78, both sexes)

ar
M
er
- Child mortality rate: Spain 1975-1997.
vi
Xa
♦ neonatal + postneonatal: O 21/1000 to 6/1000

c

♦ perinatal: O 19/1000 to 5/1000

s
ic
om
on
Ec

 Negative effects on health:

lth
ea

- Suicide and selfinjuries rate: Spain 1960-97. 4

H
of

smooth since 1975

es
pl
ci

- tobacco consumption: Spain 1960-97. 4 1000

in
Pr

ciggarettes/inhab/year. →

- Lung cancer mortality rate: 4 21/105 to 69/105

(males)
Source: Corugedo et al. (1999, p. 273-276)

49-a
 lt
F Economic development and health expenditure

ira
G
z-
ne
Positive relation and more than proportional:

ti
ar
M
er
vi
Xa
4Health expenditure
>1

c

4GDP
s
ic
om
on

F Health expenditure and effects on health

Ec
lth
ea

Ambiguous effect. Decreasing returns of Health func-

H

tion →
of
es
pl
ci

Initial stages of a health system, big impact; Mature

in
Pr

health systems modest impact.

49-b