CHAPTER ONE INTRODUCTION This chapter entails background to the study followed by statement of the problem, objective of the

study, statement of hypotheses, significance of the study, scope of the study, and organization of the study. 1.0 Background to the Study

Over the last 15 years, Ghana has become a showcase of successful macroeconomic reforms and stabilization programs. Starting with its economic reform program in 1983, Ghanas government has managed to achieve a broad budget balance, to implement a system of flexible exchange rates, and to liberalize the countrys trade regime. Also, it started to install an investor-friendly regulatory environment, initiated the privatization of 230 state owned enterprises, and began to significantly reduce the export tax on cocoa, the countrys second most important export product after gold. Ghana is often described as one of West Africas development success stories: the countrys growth and poverty reduction rates are among the very best in the region. Ghanas stable, peaceful political climate supports a policy environment conducive to economic and social progress and poverty reduction. Ghana's economy and politics are stable and should continue to grow as the sub-Saharan African nation enters its 50th year, according to a Deutsche Bank analyst. Deutsche Bank Group analyst, Marion Muhlberger, said with an annual output of 2 million ounces of gold per year, Ghana is the second-largest gold exporter on the continent, behind only South Africa. And with gold prices well above $700 an ounce, the economy should continue to perform, she said in a report published on the bank's Web site Friday said. "Solid annual economic growth of around 6 percent in real terms, inflation rates only barely into double digits, a relatively stable currency, satisfactory public finances and low foreign debt thanks to forgiveness measures were the icing on the cake," Muhlberger said in the report, originally published in the German newspaper Borsenzeitung. Also, she noted a recent discovery of at least 600 million barrels of oil off the country's coast, and while production will not happen immediately, the "black gold" will generate "considerable additional revenues for Ghana in the medium term at least," she said.

The countrys economy was the fastest growing in the world in 2011, recording growth rate of 13.4 percent. Launching the Banks global economic prospects report via a video-conference today, Mr. Andrew Burns, Manager of Global Macroeconomics in the World Banks Prospects Group, said Ghana is still in position to register strong economic growth without the oil sector, particularly in construction services as large infrastructure projects are being undertaken. He said, Ghanas economy benefited from strong rebound of both volumes and prices of gold and cocoa increase in tourism. However as the saying goes Statistics is like a bikini on a sexy woman. It shows you a lot of interesting stuff but it does not show you everything. Such is the nature of Ghanas impressive but uninspiring growth data. At 13.6%, the country is one of the fastest growing economies in the world. Last year, it was elevated to lower middle-income status after a recalculation of GDP figures. Not a bad record, except that underneath the stratospheric growth numbers lays serious troubles unemployment, poverty and poor living standards. Nonetheless, government officials and international bureaucrats alike never tire of whipping up such impressive but uninspiring economic data to justify why they think they are doing a fine job as stewards of the economy. Asked why he deserves a second term in office during his latest interaction with the Ghanaian media, President Mills was quick to cite GDP growth amongst his top three reasons. Ghana remains a star pupil of the IMF and the World Bank largely because of its impressive growth indicators. Yes, growth matters. But the fact remains that although growth rates may indicate economic boom; however, the reality on the ground in Ghana is very different. Apart from the President and his technocrats at the Bank of Ghana and the Ministry of Finance who are bustling in their glory, common folks wonder and often ask; where exactly do these figures come from? If Ghanas economy is on such an inspiring growth path, should it not be creating jobs, raising standards of living, and improving the welfare of its people? These are questions worthy of asking but yet again the Ghana Statistical Service provides another bikini figure for answers. The truth is, despite the upward growth rates, Ghanas economy has been unable to provide enough jobs for the endless stream of graduates emerging from the universities, colleges, polytechnics and the secondary schools. The cocoa industry, the backbone of the economy for the past 100 hundred years, has failed to attract new farmers, especially the educated. The mining sector is largely controlled by foreign interest. It employs
2

very few people and most of the profits are repatriated abroad. Manufacturing is largely stagnant, contributing only 9% of GDP. And oil production is unlikely to generate direct large scale employment despite the high expectations. The Ghanaian economy with its enormous resources and impressive on one side and the grim picture of economic problems presents a paradox. Consumption has remained an important parameter in evaluation of economic development. There is indeed an established relationship between human quality of life and its impact reflected through their consumption pattern. Consumption expenditure is one of the primary components of Gross Domestic Product (GDP).GDP of any nation is nothing but the aggregate demand of goods and services. The relation between aggregate income and aggregate consumption is a significant component in explaining the analysis of national Income. The effect of consumption function on economic fluctuations was mentioned by all the eminent economists from Malthus to Wicksell in explaining the business cycle literature and its contribution in economic development. Over a period every economy, whether developed or developing, has observed change in the consumption pattern, which had been an outcome of changing life style. Better quality of life is an indicator of economic development and consumption pattern has changed with escalation in quality of life, proving its significance. The aggregate consumption expenditure is one of the largest components of Gross National Product (GNP) accounting for 65-75 per cent of it in different developed and developing economies. Consumption expenditure is one of the most important macro variable, alongside investment, government spending and net export (export minus import) used in macroeconomic analysis in real terms and also the most important factor in determining the level of economic activities in an economy (Dwivedi, 2005). It is, therefore, not surprising that Ghanaians spend larger proportion of their income, largely on goods and services (Ghana Statistical Service, 2006). Such a country normally has high rate of interest, low investment and tends to grow relatively slow. This is sometimes referred to as present-oriented country. A recent literature has documented a large rise in consumption inequality in several developing countries, including Ghana. Global inequality in consumption, while reducing, is still high. Using latest figures available, in 2005, the wealthiest 20% of the world accounted for 76.6% of

total private consumption. The poorest fifth just 1.5%:

Breaking that down slightly further, the poorest 10% accounted for just 0.5% and the wealthiest 10% accounted for 59% of all the consumption:

In 1995, the inequality in consumption was wider, but the United Nations also provided some eye-opening statistics (which do not appear available, yet, for the later years) worth noting. Todays consumption is undermining the environmental resource base. It is exacerbating inequalities. And the dynamics of the consumption-poverty-inequality-environment nexus are accelerating. If the trends continue without change not redistributing from high-income to low-income consumers, not shifting from polluting to cleaner goods and production technologies, not promoting goods that empower poor producers, not shifting priority from consumption for conspicuous display to meeting basic needs todays problems of consumption and human development will worsen. According to the UN Development Programme's 1998 Human Development Report, there is nothing wrong with consumption. In fact, consumption is essential for economic growth. The real issue is not consumption itself but its patterns and effects. There is something very wrong with the great disparities in current consumption patterns. Inequalities in consumption are stark. Globally, the 20% of the worlds people in the highest-income countries account for 86% of total private consumption expenditures the poorest 20% a minuscule 1.3%. More specifically, the richest fifth:
5

Consume 45% of all meat and fish, the poorest fifth 5% Consume 58% of total energy, the poorest fifth less than 4% Have 74% of all telephone lines, the poorest fifth 1.5% Consume 84% of all paper, the poorest fifth 1.1% Own 87% of the worlds vehicle fleet, the poorest fifth less than 1%

Today two out of three Africans live on $2 per day. But things are changing. Africa is developing new consumers from developing economies, and their purchasing power today ranks among the very poor and the middle class. They earn between 2 and 10 dollars per day and have solved their existential problems. They are teachers, nurses, small businessmen, private sector employees. They are trying to improve their comfort and can afford consumer electronic goods, send their children to private schools, extend their homes and most of them have a second job. Since 1950, the world's consumption expenditure has increased six-fold, to $24 trillion in 1998. Of this, 86 per cent is accounted for by the richest fifth of the world's population, while the poorest fifth's share is only 1.3 per cent. At $200 billion, sub-Saharan Africa's 1995 consumption expenditures amounted to a mere 0.9 per cent of the $21.7 trillion global total. The average African household consumes 20 per cent less today than it did 25 years ago. The report points out that the entire world population's needs in the areas of basic health and nutrition, water and sanitation, basic education, as well as women's reproductive health could be met with additional annual expenditures totalling $40 billion. This compares with the $50 billion and $105 billion Europeans spend annually on cigarettes and alcoholic drinks, respectively, and with global spending of $400 billion and $780 billion on narcotic drugs and the military Food consumption expressed in kilocalories (kcal) per capita per day is a key variable used for measuring and evaluating the evolution of the global and regional food situation. A more appropriate term for this variable would be national average apparent food consumption since the data come from national Food Balance Sheets rather than from food consumption surveys. Analysis of FAO statistical database shows that dietary energy measured in kcals per capita per day has been steadily increasing on a worldwide basis; availability of calories per capita from the mid-1960s to the late 1990s increased globally by approximately 450 kcal per capita per day and by over 600 kcal per capita per day in developing countries (see Table 1.1). This change has not,
6

however, been equal across regions. The per capita supply of calories has remained almost stagnant in sub-Saharan Africa and has recently fallen in the countries in economic transition. In contrast, the per capita supply of energy has risen dramatically in East Asia (by almost 1000 kcal per capita per day, mainly in China) and in the Near East/North Africa region (by over 700 kcal per capita per day). Table 1.1 Actual and Projected Global and Regional per capita food Consumption (kcal per capita per day) Region World Developing countries Near East and North Africa Sub-Saharan Africa* 1964 - 1966 1974 - 1976 1984 - 1986 1997 - 1999 2015 2358 2054 2290 2058 2435 2152 2591 2079 2546 2105 1986 3065 3385 2655 2450 2953 2057 2689 2559 2205 3206 3379 2803 2681 3006 2195 2824 2921 2403 3380 2906 2940 2850 3090 2360 2980 3060 2700 3440 3060

Latin America and the Caribbean 2393 East Asia South Asia Industrialized countries Transition countries
*

1957 2017 2947 3222

Excludes South Africa.

Source: FAO, 2003. A new social class, enjoying a "little prosperity, has emerged in Africa. The emergence of the African middle class has transformed the black continent into one of the most promising markets in the world. This new class now has some twenty million people, against 12.8 million in 2000. In 2030, they will be around 43 million. They live mainly in the most highly developed countries of the continent such as South Africa, Nigeria or Ghana. The middle classes are also developing in Senegal, Ivory Coast and Kenya, and they are adopting new consumption patterns. With the development of a true middle class, consumption patterns are changing. Shopping malls and large retailers are increasing across the continent. In Ghana, a new mall has just seen the day
7

in the heart of the capital. The Accra Mall hosts over 20,000 square meters and more than 65 shops. Pan-African groups from South Africa, such as Shoprite and Game, have opened their stores there. The first group is also present in 19 sub-Saharan countries and the second in eleven. This shows that African companies are now developing across the continent. Africas growth has exceeded 5% per annum over the last decade (Only Asia has done better), and its consumption potential is not so far from that of India. Consumption fluctuates much less than Gross National Product (GNP).The least stable component of consumption is durable consumption. Services and nondurable consumption grows more smoothly. The main reason that consumption fluctuates much less than GNP is that disposable income fluctuates less than GNP. Consumption is financed out of disposable income. Over the past decades in the United States of America, consumption has more or less tracked income according to a simple Keynesian consumption function with a MPC of 0.91 of each incremental dollar of disposable income, 91 cents has been spent on consumption goods and 9 cents has been saved (Hall and Taylor, 1988). Consumption theories explain the relationship between income and consumption. It is, therefore, very crucial to have a working knowledge of the theories of consumption, determination of consumption expenditure and the nature of relationship between consumption and its determinants. Economists agree that income is the main determinant of consumption expenditure, but they differ widely on the nature of income. According to Keynes (1936), consumption depends on the current, absolute income of the consumer; Duesenberry hypothesizes that consumption depends on relative income, i.e. income of a household in relations to that of the peer group; Friedman postulates that consumption is related to permanent income, i.e. the lifetime income; Robert Ando and Franco Modigliani relate consumption to income cycle of the consumer over his or her lifetime (Dwivedi, 2005). Each of these hypotheses has its own strengths and weaknesses and relevant in a particular context. It must be noted that change in income does not fully explain the change in consumption expenditure although income is the dominant determinant of the consumption level. It implies that there are some non-income factors also which influence the proportion of income spent. As Ackley (1969) has remarked, Either there is an erratic, unexplainable elements in spending, particularly important in the short run, or there are other systematic factors influencing
8

consumption that need to be brought into the analysis. These non-income factors include interest rate, wealth, change in price level, expectations, demography etc. Some economists have made empirical studies to find the influence of such factors on the household consumption. The empirical findings have, however, not produced conclusive evidence on non-income factors in the aggregate consumption 1.1 Statement of the Problem

A nations economic performance or growth can be influenced by consumption, saving, and investment etc. Future oriented nations that save and invest large proportion of their income achieve growth of output and income. This is because such a country normally has high saving and investment ratios. Consumption and saving decisions are at the heart of both short- and longrun macroeconomic analysis (as well as much of microeconomics). In the short run, spending dynamics are of central importance for business cycle analysis and the management of monetary policy. And in the long run, aggregate saving determines the size of the aggregate capital stock, with consequences for wages, interest rates, and the standard of living. Long-term economic growth requires capital investment in infrastructure, education and technology, for example, as well as in factories, business expansion, and so forth and the main domestic source of funds for capital investment is household savings. Consistently high savings rates over time in a particular country can translate into funds being available for this long-term growth. In addition, higher levels of household savings allow a larger portion of a countrys overall debt to be financed internally. Analysts consider a high debt level primarily financed by domestic savings to be more easily sustained than high debt levels primarily financed by external (foreign) creditors Sub-Saharan Africa has the lowest savings rate in the developing world. While figures vary from country to country, gross domestic savings in the region averaged about 18 per cent of gross domestic product (GDP) in 2005, compared with 26 per cent in South Asia and nearly 43 per cent in East Asia and Pacific countries, according to World Bank estimates. In some countries, those rates are even on the decline There are many reasons for Africas low savings rates, including inadequate financial services African countries ability to finance a greater share of their development needs from domestic sources would give them much-needed flexibility in the formulation and implementation of
9

policies to address development challenges, direct resources into high-priority areas and strengthen state capacity, finds a 2007 UNCTAD Yet until recently, most international conferences and summit meetings to address the financing of Africas social and economic development have generally focused on ways to mobilize more foreign resources. That is changing. But flows of official development assistance (ODA) to Africa remain volatile, and as the UNCTAD report notes, dependence on external resource flows leaves countries vulnerable to external shocks. Moreover, the regions share of global foreign direct investment (FDI) has stayed low. As a result, African governments are increasingly turning their attention to the need to enhance savings rates. In Ghana, for instance, since the early 1960s, consumption has risen substantially as a share of Gross Domestic Product (GDP). This might undermine the motive of stabilizing inflation by putting upward pressure on prices of domestic commodities. This also partly influences the slow growth rate of the Ghanaian economy. According to the Ghana Statistical Service (2006), the average annual household income in Ghana is about GH1,217.00 whilst the average per capita income is almost GH400. There are regional differences with Greater Accra region recording the highest of GH544.00 whilst Upper West and Upper East regions had less than GH130.00. Urban localities had higher per capita income than rural localities. The three main sources of household income in Ghana are income from agricultural activities (35%), wage income from employment (29%) and income from selfemployment (25%). Remittances constitute less than 10 per cent of household income. The annual estimated total value of remittances received in Ghana is GH547,571 million whilst the estimated total annual value of remittances paid out by households is GH231,344 million which represents 42 per cent of all remittances received. In addition, the Ghana Statistical Service (2006) revealed that the average annual household expenditure in Ghana is GH1,918.00 whilst the mean annual per capita consumption expenditure in Ghana is GH644.00. Regional differences exist with Greater Accra Region having the highest per capita expenditure of GH1,050.00 whilst Upper West has the lowest of GH166.00. The average annual household expenditure is about 1.6 times higher in urban localities (GH2,449) than in rural localities (GH1,514) even though the household size in rural
10

households tends to be larger than urban households. Food expenditure accounts for two-fifth of total household expenditure, while the imputed value of own-produced food consumed by households represents a further 10.5 per cent. Expenditure on housing in Ghana averages 2.4 per cent of the total household expenditure. Expenditure on housing is higher in Greater Accra Region than the other regions. At the time of the survey Ghanaian households were spending on average an amount of almost GH2,680 million per annum with food (including non-alcoholic beverages) representing about a third of the total expenditure while non-food expenditure represented about 70 per cent of the total household expenditure. Within the non-food expenditure group, transport contributes the highest of 16.7 per cent to the total expenditure. The next most important expenditure groups in terms of amount spent are housing, water, electricity and gas (7.9%), recreation and culture (6.1%) and education (5.3%). A negative savings rate indicates that a household spends more than it receives as regular income and finances some of the expenditure through credit (increasing debt), through gains arising from the sale of assets (financial or non-financial), or by running down cash and deposits. One important rhetorical question that readily follows is whether or not a country whose average annual household expenditure is greater than its average annual household income (as in Ghana) can make any significant investment in order to achieve rapid economic growth and development? The sprawling Obuasi township consists of three faces: the plush residential area for Anglogold Ashanti ( AGA) management, AGA junior staff residence, and the main Obuasi town, which from the point of view of visitors has very little resemblance to a resource-rich town. Wide consumption inequality exists among households of these sections of the town. Consumption is normally high among households in the plush residential area for AGA management. For households in the AGA junior staff residence, consumption is abnormally high. Indeed in Obuasi, gold mining jobs are somewhat the most widely available and highest paying. Sadly, you do not need to have experience, a clear criminal background, or even a high school diploma, just common sense, a strong work ethic, and a clean drug test. Work hard, play hard is the motto as miners spend extravagantly to enjoy what little time they have off. Most of the men are deeply in debt after buying new homes, vehicles, LCDs. Some even plan what payments they can afford based on how much overtime they plan on working. It is a vicious
11

cycle that keeps coal miners scared for their job for fear of losing everything they have, working to the advantage of the gold companys production. On the contrary, consumption rate is very low among households of the main Obuasi town. Residents here are mainly farmers from the surrounding communities of Binesere, Dokyiwa, Mamiriwa, Sansu, Anyinam, Apitiso, Nhyiaeso, Anwiam among many others who have been rendered unemployed due to the operations of the mining company. They are mainly petty traders and manual labourers. A thorough observation of the main Obuasi town reveal sporadic consumption rate among the folks due to fluctuations in their incomes. That is, household consumption is very low at one point in time and very high at another period. This menace stems from the fact that petty trading mainly in foodstuff is the dominant occupation among the people of the town; therefore, consumption is expected to be high during the bumper harvest especially around August when income of both farmers and traders is expected to rise astronomically and fluctuates, thereafter. A research conducted on behalf of the municipality by some NGOs in 2008 disclosed that most households spend extravagantly during occasions such as funerals (especially the purchase of different type of funeral cloth). Some parents are adamant and reluctant to pay for their childrens school fees which adversely affect those children in their later life. Although measures were put in place to redirect household spending to areas such as education, this had not yielded much result. It is in view of this problem that the researcher intends to determine what factors actually influence the level of household consumption in Obuasi Township. Is it only income or income together with other socio-economic factors such as wealth, interest rate, change in price level, expectations, demography etc.? 1.2 Objectives of the Study

1.2.1 General Objective To determine the factors that influence household consumption 1.2.2 Specific Objectives 1. To determine the level of household income and their effects on consumption.

12

2. To examine the relationship between household consumption and socio-economic variables such as income, family size, age, wealth, remittance and education. 3. To verify and investigate Keynes conjecture of fundamental psychological law. 4. To suggest some policy recommendations

1.3

Statement of Hypotheses

In relation to the statement of the problem and the objectives of the study, the following hypotheses are formulated for testing. Ho: Household income does not influence consumption, against H1: Household income does influence consumption. H0: Household income, family size, age, wealth, remittances, and education jointly influence the level of household consumption, against H1: Household income, family size, age, wealth, and remittances, and education do not jointly influence the level of household consumption

1.5

Significance of the Study

The rationale of the study is to analyze the consumption pattern of the people of Obuasi and explore issues regarding the factors that influence consumption. There is no doubt that aggregate consumption is a key variable for policy makers. The aim of this research is to determine factors that significantly determine household consumption and draw out the implications for policy analysis. This finding is intended to be accessible to those working in policy-related departments without losing economic rigour. Specifically, this research will serve as a guide to the government, policy makers and the council of elders of Obuasi town on their decisions on how to improve the consumption pattern of the people of Obuasi and Ghanaians as a whole and on the flip side, how to promote household saving which is essential for sustainable growth and development given that dependence on external resource flows leaves countries vulnerable to

13

external shocks. The findings may also serve as guide and reference material to the general public, students and researchers who are researching into similar areas 1.6 Scope of the Study

The focus of the study is to examine the determinants of household consumption. However the study has been narrowed to Obuasi town in the Ashanti Region. This stems from the fact that not much of such research has been done in the area. To ensure easy access to the relevant information, the researcher segmented Obuasi into three. These are the residential area for Anglo-gold Ashanti ( AGA) management, AGA junior staff residence, and the main Obuasi town. The multiple linear regression model will be used to establish the significance of the individual explanatory variables on consumption. 1.7 Organization of the Study

The study is divided into five chapters as follows: Chapter one covers the background of the study followed by statement of the problem, objective of the study, statement of hypothesis, significance of the study, scope of the study and organization of the study (which gives an insight into how the study has been organized). Chapter two deals with the review of literature for the study. The literature review comprises theoretical literature and empirical literature review. Chapter three is concerned with the research methodology used in the study. It deals with the specification of the model for the study, research design, population and sampling techniques, research instrument, data source and method of data collection, definition and measurement of variables, and estimation techniques. Chapter four is devoted to the presentation, analysis and discussion of the results. It also deals with the testing of the hypotheses formulated in the first chapter. Chapter five, being the last chapter, contains the summary, conclusions, policy implications, limitations of the research and areas for future study

14

CHAPTER TWO LITERATURE REVIEW 2.0 Introduction

This chapter is divided into four sections. The first and second sections deal with the review of theoretical literature and empirical literature respectively of renowned economists on consumption. The third section deals with the policy implication of the income-consumption hypothesis and last section gives concluding remarks of this chapter. There is no topic in macroeconomics that has a longer, deeper, or more prominent literature than households choice of how much of their income to consume and save. Economists have long been interested in the factors determining how a society divides its income proportionally between consumption and saving. In the past thirty years theoretical and empirical investigation of these factors has been focused by the concept of the consumption functiona list of the variables that influence consumption, together with the direction and magnitude of their effects. Income itself is, of course, high on any such list; and much of recent investigation has concerned the nature, reliability, and measurement of the dependence of consumption on income.

2.1

Theoretical Literature Review

The term consumption is the using up of the services yielded by a good, while the act of purchasing it is referred to as consumer expenditure. Goods vary in the length of time over which they yield services. This comprises durable goods and non-durable goods. Durable goods (for instance, expenditure on television sets, refrigerators, computers etc.) yield services over a long period, say more than one year. On the other hand, non-durable goods (for example, expenditure on food, water etc.) provide services over a relatively short period, say less than one year. Consumption has generally been defined as the value of non-durables plus the value of flow of services of durable. (Department of Economics, 2008) Keynes (1936) was the first to develop a systematic theory of aggregate consumption spending by the household in his General Theory of Employment, Interest and Money. The Keynesian theory of consumption in its rudimentary form has already been highlighted in chapter one.
15

Keynes theory of consumption was, however, challenged after the Second World War on the ground that the household consumption depends not only on current income but also on a number of other factors viz. real wealth, interest rate , taxation, availability of consumer credit, consumers expectations and income distribution so far as aggregate consumption is concerned. In the process of debate on the issue as to what determines the level of consumption; some significant contributions were made to the theory of consumption. The development in this area can be attributed to the lack of empirical evidence to support the various hypotheses that were developed by the economists. However, the economists generally agree that the household consumption expenditure is a function of household income but they are not unanimous on which income- absolute or relative income, current or expected future income, short-run or permanent (long-run income) or income-cycle over life-time. . The absoluteincome hypothesis is linked to the basic principle of Keynes theory of consumption. Keynes General Theory of Employment, Interest and Money was published in 1936.Keynes viewed consumption expenditures in the aggregate as being primarily dependent on current net income. He contended that, as net income increased, consumption expenditure would increase but by less than the increase in income. In other words, some savings will take place. If the marginal propensity to consume is defined as the change in consumption which is associated with some small change in income, then this assumption imposes the restriction that the marginal propensity to consume(MPC) will lie between zero and unity. As well as imposing this restriction Keynes made two other behavioural assumptions: 1. The first was that at low (but unspecified) income levels, consumption will exceed income, expenditure being financed by dissaving. This amounts to saying that the marginal propensity to consume is less than the average propensity to consume (APC).The latter is defined as the ratio of consumption expenditure to income(C/Y). 2. The second assumption was that the marginal propensity to consume (C/Y) or in the limit dC / dY will itself decline as income increases.

The implications of this assumption are worth emphasizing. First, it is clear that if information on the MPC exists, it is possible to predict how consumption expenditure will change in response to a change in disposable income (effected for example by a change in tax rates). Second, it implies that the percentage of income saved increases as society becomes richer. This in turn
16

suggests that in a growing economy a greater proportion of investment will be required to maintain full employment income levels. Third it suggests that a transfer of income from highincome to low-income consumers will raise the level of aggregate demand, since MPCs differ between the two groups .Thus implicit in the Keynesian approach is a justification of progressive taxation as a basis of income redistribution in conditions of surplus capacity. The Keynesian consumption function provided considerable impetus to empirical investigation. Early studies employing cross-section budget data appeared to support Keyness intuition in almost every detail. Such cross-section data related the level of consumption to income levels for different income groups, and in so doing tended to generate consumption functions remarkably similar to that of Keynes giving a linear functional form. Although the cross-section data lent some support to the Keynesian thesis, a number of early time series(that is, studies relating observation on consumption and income for specific groups through time) raised fundamental questions. Kuznets (1946) classic study for example suggested that consumption was a linear function of disposable income with a constant marginal propensity to consume of approximately 0.9. This and other time series evidence implied that savings would be a constant rather than increasing proportion of national income, thus combating the pessimistic leanings towards stagnation. Furthermore it questioned whether redistribution from upper to lower-income groups would have any impact upon the level on aggregate demand . This apparent conflict in empirical findings provided a challenge to the generality of the absolute income hypothesis and an impetus to further research directed at providing some plausible explanation of why cross-section and time series results should offer such conflicting views of the consumption function. One of the earliest influential treatises which sought to modify the Keynesian consumption function, and did so in such a way as to offer an explanation of the conflicting cross-section and time series findings, was advanced by Duesenberry (1952). He argued that Keynesian analysis focused on an inappropriate concept of income; he contended that, when explaining variations in consumption, the relevant concept of income was not absolute income but relative income. By relative income he had in mind the current disposable income of a consuming unit relative to previous peak income, and current disposable income of that same consuming unit relative to the current disposable income of other consuming units. The former of these two relativities is
17

responsible for the so-called ratchet effect, that is, the idea that the consumption-income relationship may be asymmetrical in so far as there are downward rigidities to the reduction of consumption expenditure. The latter relativity has become known as the demonstration effect and refers to the possibility that consumption decisions may be interdependent rather than independent. This phenomenon is also known as keeping up with the Joneses. If both of these effects are taken into consideration , it is possible to reconcile the findings of cross-section and time series evidence. The relative income hypothesis can be presented in the form of the following four propositions. In other words, to present clearly the four propositions of the relative income hypothesis, we single out a household H from a group of households with more or less the same level of income and analyze its consumption behaviour in response to change in its income in relation to the income of other households. If income of all households belonging to the group increases, then the consumption level of all households of the group goes up at the same rate and vice versa. That is, (MPC)

remains the same for all the households if their income changes by the same amount. If household H remains at the same scale of relative income and its absolute income rises, then its absolute consumption and savings rise, but its MPC remains the same as it was before the rise in its income. If household H remains at the same scale of relative income (with income constant) and income of the other households of the group increases, then MPC of the household H with constant income increases. If household H moves up from a lower income-group to a higher income-group then its MPC decreases. The last proposition supports the fourth property of Keynes absolute income hypothesis. Other propositions of the relative income hypothesis make a significant deviation from the absolute income hypothesis. Furthermore, both hypotheses have a similar view that when absolute disposable income of households increases, their relative income remaining the same, then consumption increases such that households average propensity to consume (APC) remain constant and is equal to the MPC. However, the relative income hypothesis deviates from the absolute income hypothesis on the question as to what happens when household income
18

decreases. While Keynes holds that consumption decreases in proportion to decrease in income, Duesenberry holds that consumption does not decrease in proportion to decrease in income because of what he (Duesenberry) calls the Ratchet Effect. The ratchet effect arises due to households resistance against the fall in consumption following a decrease in household income. Duesenberry argues that when absolute income increases, absolute consumption increases but when absolute income decreases, the households do not allow their consumption to fall in proportion to the fall in their incomes. It is so because households get used to a certain standard of living in the long run and hence when their income falls, their consumption falls less than proportionately. When consumption does not fall in proportion to the fall in income, then APC rises and MPC falls. This is called the ratchet effect in consumption behaviour. The relative income hypothesis is illustrated in fig. 2.1. Let the long run consumption function be given by the line CL = bY. Given the consumption function, suppose that, at a point of time in the long-run, households have an income equal to 0Y2 out of which they consume CY2. At this level of income and consumption, APC = CY2 / 0Y2. Now let the household income decrease in the short-run to 0Y1. According to the absolute income hypothesis, the consumption would fall to MY1 and APC = MY1 / 0Y1. In that case, APC will remain the same, i.e., MY1 / 0Y1 = CY2 / 0Y2. It implies that the fall in the household consumption due to the fall in income would be proportional. According to the relative income hypothesis, however, the decrease in the household consumption would be less than proportional because households resist the decrease in the standards of living when there is a short- run decrease in their income. Therefore, their consumption decreases but less than proportionately. This point is illustrated by the line CSC in fig. 2.1. When income decrease from 0Y2 to 0Y1, the household consumption decreases, say, to NY1, not MY1. Note that, according to the relative income hypothesis, the fall in household consumption due to the fall in income by Y1Y2 is lower. Given the CL = bY schedule, a fall in income by Y1Y2 should have caused a decline in consumption by BM whereas it decreases by only BN. Note that BN < BM. It means that ratchet effect causes a lower fall in consumption than expected.

19

The ratchet effect keeps the consumption at point N. When we join point N with point C, the resulting line NC or CSC gives the short-run consumption function. Note also that AN is the amount of dissaving. It implies that when household income falls, the households resort to dissaving in order to prevent a large fall in their living standards. They do so to maintain their living standards on par with their peer groups. This is an important point of distinction between the absolute and relative income hypothesis. Furthermore, it can be seen in Fig 9.2 that short-run average propensity to consume (APCS) is greater than long-run average propensity to consume (APCL). At point N on the short-run consumption function (CSC), APCS = NY1 / 0Y1 and at point M on the long-run consumption function (0C), APCL = MY1 / OY1. As the figure shows, NY1 > MY1. Therefore, NY1 / OY1 > MY1 / OY1. This proves that short-run average propensity to consume is greater than the long-run average propensity to consume

Consumption C=Y CL=bY C2 B C

CS

A C1 M

45o O Y1 Y2 Disposable Income Fig. 2.1 Relative Income Hypothesis

20

Source: Dwivedi, 2005 Like absolute income hypothesis, the relative income hypothesis is also marred by its limitations, though not significant enough to pose a serious challenge to the validity of the theory. One, the relative income hypothesis states that an upward change in income and consumption is always proportional irrespective of whether the change in income is small or large. The empirical evidence, however, suggest that exceptionally large and unexpected increases in income are often associated, at least initially, with less than proportionate increase in consumption. Two, the relative income hypothesis states that consumption standards are irreversible. It, however, argued that this proposition may hold in the short-run but not in the long-run. People cannot go on dissaving in the long run in order to maintain the living standards at a level higher than sustainable under the condition of decreasing income. That is, the consumption standard is reversible in the long run. This criticism is, however, not very relevant because the relative income hypothesis does not admit the reversibility of consumption expenditure with decrease in income but less than proportionately. That is, reversibility argument of the critics matters only with regards to proportionality. Three, the relative income hypothesis states that income and consumption change always in the same direction. It implies that recession must always be accompanied by a fall in aggregate consumption expenditure. There have, however, been instances that during the 1948-49 in the US, consumption expenditure was rising while disposable income was decreasing. Obviously, income and consumption had changed in opposite direction. Such exceptions, however, do not reduce theoretical importance of the relative income hypothesis because such cases are rare across the households. It may, thus, be concluded that despite its criticism based on some minor empirical aberrations, Duesenberrys relative income hypothesis is regarded as a significant improvement over the absolute income hypothesis as it resolves certain paradoxes of the absolute income hypothesis. The absolute income hypothesis relates household consumption to the current absolute income and the relative income hypothesis relates it to the current relative income. Both these hypotheses relate consumption to current income-absolute or relative. Milton Friedman rejected the current income hypotheses and developed another theory on consumption, popularly known as permanent income hypothesis. According to the permanent income hypothesis, it is the
21

permanent income, not the current income which determines the level of current consumption expenditure. Permanent income, defined broadly, is the mean of all the incomes anticipated in the long-run. The method of estimating permanent income is an approximation of incomes anticipated from all human and non-human wealth (or capital). In simple words, it means labour income plus capital incomes. If all material, financial and human sources of income are treated as wealth, then the permanent income of the current year can be defined as YP = rW, Where YP is the permanent disposable income with reference to the current year, W represents overall wealth and r is the rate of return. Friedman emphasizes that people experience random and temporary changes in their incomes from year to year. He calls that part of income that people expect to persist into the future as permanent income and the other part which people do not expect to persist into the future as transitory income. The basic proposition and assumption of the permanent income hypothesis can be stated algebraically as follows. CP = kYP. Thus, permanent consumption (CP) equals k proportion of permanent income (YP). YM = YP + YT. Thus, measured income (YM) equals permanent income (YP) plus transitory income (YT). CM = CP + CT. Thus, measured consumption (CM) equals permanent consumption (CP) plus transitory consumption (CT). R1 (Y1t, Yp) = 0: Correlation coefficient (R1) between Y1t and Yp equals zero. R2 (C1t, Cp) = 0: Correlation coefficient (R2) between C1t and Cp equals zero. R3 (Y1t, C1t) = 0: Correlation coefficient (R3) between Y1t and C1t equals zero. Households with high permanent income have proportionally higher consumption.

The equations are self-explanatory. Yet the first equation needs some elaboration. It states that permanent or planned consumption is a certain proportion (k) of the permanent income. The

22

proportionality factor k needs not be a constant for it depends on demographic and ethnic factors, interest rate, and the ratio of non-human wealth to permanent income.. Finally, the life-cycle theory of consumption, popularly known as life-cycle hypothesis was developed by Albert Ando and Franco Modigliani in the early 1960s. Like Friedmans permanent income hypothesis, the life-cycle hypothesis too rejects the Keynesian consumption theory that the current consumption depends on the current income. The life-cycle hypothesis postulates that individual consumption in any time period depends on resources available to the individual, the rate of return on his capital, and the age of the individual. The resources available to the individual consist of his existing net wealth and the present value of all his current and future labour income. According to the life-cycle hypothesis, a rational consumer plans consumption on the basis of all his resources and allocates income to consumption over time so that he maximizes his total utility over his life-time. The basic propositions of the life-cycle hypothesis can be summarized as follows. The total consumption of a typical individual depends on current physical and financial wealth and his life-time labour income. Consumption expenditure is financed out of the life-time income and accumulated wealth. The consumption level of a typical individual is, more or less, constant over his lifetime. There is little connection between current income and current consumption.

The first and second propositions can be transformed into a life-time consumption function as follows. C = aWR + cYL Where WR is real wealth, YL is labour income, a is MPC wealth income, and c is MPC labour income. To explain the life-cycle hypothesis, let us suppose that an individual expects to live for N years with his retirement age at R. He starts working at the age of B, i.e. his working life equals R B years. For simplicity sake, we assume also: (i) that the individual has no uncertainty about his
23

longevity, employment and health condition; (ii) that he earns no interest on his accumulated savings; (iii) that he does not consume his total labour income; and (iv) that prices remain constant. With these assumptions, his life-time income is estimated as follows. Life-time Income = YL (R - B) Where YL is annual labour income, and (R B) is number of working life. According to the life-cycle hypothesis, an individual plans his life-time consumption in such a way that his life-time consumption equals his life-time income. Here the term life-time means working life. Given the individuals expected life of N years and his planned constant (annual) consumption (C), the consumption hypothesis can be written as: C X N = YL X EL By dividing both sides by N, we get the life-time consumption (C) as C = EL / N X EL / N Income/Consumption

WEALTH

INCOME Y C SAVING CONSUMPTION 0 DISSAVING Retirement begins Fig. 2.1: End of years Years

Life-Cycle Hypothesis

24

The figure 2.1 above shows the life-cycle hypothesis. If the consumer smooth consumption over his/her life (as indicated by the horizontal consumption green line), he/she will save and accumulate wealth during his/her working years, and then dissave and run down her wealth during retirement. Like all other income-consumption theories, the life-cycle hypothesis too has its own weaknesses. First, the life-cycle hypothesis has strongly been criticized for its strong assumptions. The theory assumes that an individual has a definite vision of future size of his income, the entire profile of his life-time income, availability of present and future credits, future emergencies, opportunities and social pressures, present and future rates of interest and returns on investment, and that he has a finely planned life. These assumptions are questionable. Second, the life-cycle hypothesis assumes that the vision of the spending units right or wrong, correct or incorrect has a high degree of certainty. The worlds experience is full of uncertainty in economic life. Therefore, this kind of assumption is highly untenable. Third, this theory assume that each individual has all the information he needs; can make all the fine and complex calculation; mix rational decisions; and plans his present and future consumption so finely that it can be repeated year after year. This is an unrealistic assumption. Finally and more importantly, the empirical studies that have been carried to verify the life-cycle hypothesis do not produce supporting evidence. Instead, most studies produce evidence contrary to the life-cycle theory propositions. 2.2 Review of Empirical Literature

Various empirical studies can be found in these areas which are reviewed below. Altonji and Villanueva (2007) report that the propensity to bequeath increases with lifetime income. Agents care, not only about permanent income, but also about relative income which provides straightforward explanation for this evidence in terms of interpersonal comparisons.

Erlandsen and Nymoen (2006) tested for the age structure effects on aggregate consumption in Norway, by estimating a consumption function which takes account of changes in the age distribution of the population. Their analysis was based on aggregate time-series data over the period 1969-2004. They found that changes in the age structure of the population have
25

significant and numerically important effect on Norwegian aggregate consumption. More specifically, they found that aggregate consumption decreases when the share of the middle aged, from (50-60 years) persons in the population increases. Their results, hence, give support to the life-cycle model. Their findings implied that the changes in the age distribution of the Norwegian population may have significantly affected domestic demand and households savings. Raj (2005) used a cross-section to test the evidence on the absolute income hypothesis and the permanent income hypothesis. The key objective of this research was to verify the stylized fact that higher income households tend to save larger fraction of their income than lower income households. From the 2003 household income and expenditure survey, the result obtained were ambiguous and did not conform to the conventional theory in the strict sense. There was, however, some indication that the lower income households were consuming higher proportion of their income than the high income households. Dynan (2004) find a strong positive relationship between saving rates and lifetime income. Specifically, recent work by Dynan (2004) and Altonji and Villanueva (2007) provide strong evidence of a saving rate that increases with permanent income, violating the proportionality hypothesis.

Tan and Voss (2000) examined the relationship between private consumption expenditure and wealth in Australia for the late 1980s and 1990s. For this period they identified a steady state relationship between non-durable consumption, labour income and household wealth. Based upon this relationship an increase in aggregate per capita income of one dollar was eventually associated with a rise in annual non-durable consumption of approximately four cent. This was somewhat smaller than other estimates but not inconsistent with economic theory. They also estimated a short-run model of consumption and found that changes to household net nonfinancial and financial wealth are an important determinant of consumption growth throughout the 1990s particularly in recent years. Raid (2000) investigated the main determinant of private consumption spending for the period 1969-1992. The econometric result showed that private consumption behaviour is strongly affected by the current disposable income, previous level of disposable income, previous level of
26

consumption and the changes in the current disposable income. The paper showed that, the current domestic income, previous consumption spending, inflation, workers remittances, and the political and economic instability are the main determinant of private consumption of Jordan during the study. Friedman (1957) and Branson (1972) investigated Keynes theory using a cross-section data. They surveyed households and collected data on consumption and income at a point in time and estimated the consumption function. The study supported Keynes prediction that MPC lies between zero and one. They found that, current consumption expenditure was highly correlated with income. Moreover, when the rich and the poor households were compared with each other at one moment in time, the proportion of income saved increased along with income. One empirical study by Hamburger (1955) reveals that the ratio of wealth to income is closely correlated with the ratio of consumption to income, as judged by aggregate time-series data for the inter war and post- World War II period. Tobin (1951) made a comparative study of the absolute income and relative income hypotheses based on four different sets of empirical data to find their empirical validity. His results do not tell categorically which of these hypotheses has greater empirical validity. However, his findings based on short-run budget, support absolute income hypothesis but not the long-run relation between consumption and income. His own study was, however, been questioned too. In addition, Tobin (1951) has recently examined the consistency of the relative income hypothesis and the earlier absolute income hypothesis with a limited body of empirical evidence. Though he finds neither hypothesis entirely satisfactory, he concludes that the weight of evidence favours absolute income hypothesis and he tentatively suggest that changes in wealth may explain the rough constancy over time in the fraction of income saved. Kuznets (1946) used a trend data from 1869 to 1938 and discovered that the ratio of consumption to income was remarkably stable from decade to decade despite large increases of the income over the period. He obtained a long-run consumption function such that the long-run APC and MPC were both constant. In other words, he discovered that APC was constant implying that APC approximates MPC.

27

2.3

Policy Implications of the Consumption Theories

Policy implications of the consumption theories matter because the government is often required to influence household consumption with the view to achieve some economic goals. For instance, the government is often required to curb the household consumption during the period of high inflation, by hike in tax rate, especially when rise in price is caused by increasing consumer demand. Similarly, during the period of economic recession, the government is required as a matter of policy to revive the economy by encouraging demand by such budgetary measures as tax-cut and subsidy. The question that we attempt to answer is: how do economists view the effectiveness of policy formulation based on various consumption theories? The permanent income hypothesis is considered to be a better policy guide than the absolute and relative income hypotheses. The argument runs as follows: the absolute income hypothesis states that current consumption depends on current absolute income while according to the relative income hypothesis current consumption depends on current relative income. Therefore, the policy implication of these hypotheses is that a tax rebate and subsidy would affect consumption in the current year and the magnitude of the effect will depend on the MPC and the multiplier. According to permanent income hypothesis, however, household consumption depends on the permanent income. So the effect of a tax rebate or subsidy on the household consumption depends on how the households view the policy measure. If they view a tax rebate or subsidy as transitory gains, their consumption remains unaffected. If they expect the tax rebate or subsidy as a permanent gain, then, given the method of measuring permanent income, such policy measures would be effective immediately. The policy measures affect consumption, if they do at all, only marginally and their effect is distributed over a number of years. These conclusions have been verified with respect to the US economy by Modigliani and Steindel (1977).

2.4

Conclusion

The question that arises now is: what is the final status of the consumption theories? An answer to the question lies in the empirical validity of the consumption theories. But the empirical evidence on income and consumption relation is ambiguous.
28

According to Ackley (1978), although permanent income hypothesis and life-cycle hypothesis have earned a greater honour in recent years, for providing theoretical basis of empirical research on consumption behaviour, this seems a victory by default. The honour accorded to these theories because there is absence of serious competition and not because they have improved our understanding of aggregate consumption. It is generally agreed that short-run consumption function takes the form as C = a + bY (where Y is current income) and the long-run consumption function as C = kYP (where YP is current income. There is, however, no conclusive evidence for either form of these consumption functions. In a nutshell, it is amply shown from the literature that income is the most important determinant of household consumption. Other determinants identified in the literature include wealth, age, workers remittances and others. This study is guided by these factors when formulating the model.

29