Angus Deaton, Nobel laureate

Orazio Attanasio 23 October 2015

a

A

Since the 2008 financial crisis and the subsequent Great Recession, the economic profession has been heavily criticised. Economists were not able to predict the events that took place after Lehman Brothers folded – and they were divided about the right response to the crisis. Since then in much of the media (both traditional and new), economists are often described at best as delusional types, living in a fantasy world described by abstract mathematical models where individuals are hyper-rational, and where markets are perfect and deliver the best possible outcomes. At worst, they are accused of defending the interests of greedy capitalists and multinationals.

The life’s work of Angus Deaton, which has earned him the Nobel Prize in Economics, is a demonstration of how the best economists cannot be caricatured in this way. His research has always been driven by real world problems of huge relevance for economic policy, development and progress and – at the same time –grounded in economic theory. His work has shown what economics and an economist can offer to the policy debate and to social sciences in general. His approach has never shied away from difficult problems and it has always stressed the importance of bringing economic theory to data. He has shown that economic models and economists’ insights are useful to the extent to which they can be brought to bear on real data.

Throughout his career, he has been the opposite of the insular economist, immersed in the abstract world of perfect markets and hyper-rational consumers. On the contrary, he has used insights from other disciplines and engaged in conversations and occasional collaborations with psychologists (such as Daniel Kahneman), epidemiologists (such as Michael Marmot) and philosophers (such as Nancy Cartwright). Finally, he has not shied away from taking controversial positions, which, however, were always grounded on research rather than ideology and/or political correctness.

Demand analysis

In the first phase of his career, Deaton’s work focused on the analysis of consumers’ expenditure decisions. While the characterisation of demand for specific commodities is one of the earliest examples of applied economics, Deaton’s work constituted fundamental progress in this respect.

The theory of consumer behaviour has been established for some time and goes back to the work of Marshall, Hicks, Gorman and others. The demand for a given commodity is understood to depend on resources, the price of the commodity and the price of other commodities. In addition, it is likely that the demand for certain commodity is affected by other variables, ranging from demographics to individually heterogeneous tastes and needs. The issue is to find empirical specifications that are able to capture the richness and the complexity of the real world and still be ‘theory consistent’.

Deaton’s work in this respect has been path-breaking. Starting with his article published in Econometrica in 1978 (which was awarded the first Frisch prize for the best applied paper published in that journal over a four year period) and his American Econoic Review paper with John Muellbauer published in 1980, Deaton has shown how to construct models that can be easily estimated with survey data, can fit those data reasonably well and yield results that are consistent with a well specified model of consumer behaviour. It is for this reason, that the Almost Ideal Demand System model (known as the ‘AIDS’ model) that Deaton and Muelbauer proposed 35 years ago and its generalisations (such as ‘Q-AIDS’) are still widely used in practice.

Why is consistency with theory important? Why wouldn't one simply estimate a regression model that fits well a set of available data? The analysis of demand is relevant because it can be used to assess the effects of changes in prices or indirect taxes or even taxation on welfare and wellbeing and on its distribution in the population. However, only empirical relationships that are consistent with a well specified model of individual behaviour can be used to assess the welfare implications of certain policies or changes to prices or measuring poverty and so on.

Indeed, demand analysis and its empirical use are central to the very construction of ‘ideal’ price indexes (or to judge how closely certain price indexes approximate such an ideal). People have become very used to ‘inflation adjustments’ and to the conversion of ‘nominal’ to ‘real’ figures. And yet behind the construction of price indexes there is a theory of demand.

Saving and intertemporal choices

The next set of important contributions in Deaton’s portfolio are those that consider consumers’ allocation of resources over time. Since the 1950s, with the introduction of Friedman’s permanent income model and Modigliani’s life cycle model (see Friedman 1957, and Modigliani and Brumberg 1954), it has been recognised that when choosing how much to consume and how much to save, individuals react not only to current income but also to expectations of future income interest rates, to the extent they capture the relative prices relative price of present and future consumption. Indeed, the main intuition of Friedman and Modigliani’s models is that present and future consumption can be analysed as different commodities, the demand for which will then depend on the total amount of resources and relative prices.

There are, of course, some important differences between static demand theory and the theory of dynamic allocation of consumption over time. When dealing with the latter one cannot avoid dealing with uncertainty and the role that risk aversion might play. Preferences about the future can also be tricky, depending on how we model patience. And last but not least, moving resources through time to finance consumption might be difficult, especially when resources are in the future.

Deaton wrote some important papers on these issues. First, he recognised the importance of working with micro data even when analysing intertemporal models, which had often been studied with aggregate national account data. For this to be possible even when longitudinal data are not available, he developed techniques to use time series of repeated cross-sections for this purpose. In his paper with Margaret Irish and Martin Browning published in Econometrica in 1985, he used the clever idea of following groups rather than individuals to identify the relations of interest.

Second, he stressed the relationship between the properties of the process that generates income and its relationship with consumption – if innovations to income are transitory, they should not be reflected in consumption, while if they are of a permanent nature, consumption should adjust fully or almost fully to them (on this see his paper with John Campbell published in the Review of Economic Studies in 1989).

Third, he pioneered the use of numerical techniques for the analysis of life cycle models, a practice that has become standard (see his 1991 Econometrica paper). In that paper, he focussed on the role of precautionary savings.  He also pointed out the implications of standard models (and certain properties of the income process) for the evolution of consumption inequality over the life cycle. His paper in the Journal of Political Economy with Christina Paxson is a very elegant and convincing piece of work in this respect.

Development economics – demand, measurement, health and happiness

Deaton’s interest in consumer behaviour led him to a keen interest and many important contributions in development economics, notably on the demand for food and nutrients (e.g. Deaton 1989, Deaton and Paxson 1998). Many of these papers are based on analysing the mapping between theoretical restrictions and patterns in the data, with the identification, in some cases, of important puzzles (such as that in Deaton and Paxson 1998 on the relationship between household size, income and the demand for food) that have stimulated many other papers and studies. They are all based on solid demand analysis.

Perhaps not surprising for a student of Sir Richard Stone, Deaton has also been always interested in measurement issues, especially in that part of his research that has considered prices and price measurements (and comparisons of prices across countries) as well as in his work on measuring poverty lines and the prevalence of poverty in developing countries. His work on Indian surveys has been exemplary as has his work on adjustments for purchasing power parities. Again, these works are  examples of fundamentally important empirical work that is solidly grounded in the theory of demand. Stemming from his interest in measurement, his contributions in this area, such as his book on the design of surveys in developing countries (Deaton 1997), and his work at the World Bank, have been enormously influential.

More recently, Deaton has added two important areas to his research. First is the study of health – the so-called health gradient – and, more generally, the determinants of health during the process of development. His work, while taking on board many insights uncovered by epidemiologists, has recognised the complexity of the relationship between health and economic wellbeing and the possibility of causality running in both directions. Some of this work, started a few years ago (see Deaton 2001) has been effectively summarised in his popular book The Great Escape (2014).

Second, he has been analysing data on life satisfaction and ‘happiness’, working closely with the producers of large surveys that attempt to measure such concepts and with psychologists (most noticeably Kahneman – see Kahneman and Deaton 2010) to go beyond the measurement of material resources.

Controversies

As mentioned above, Deaton has not shied away from expressing forcefully controversial opinions. Those who have seen him in seminars and conferences know that he can be provocative and scathing in his criticisms. However, it is also clear that his positions are always based on a deep understanding of the relevant theory (be it a statistical or economic issue), a good knowledge of empirical evidence and a wide knowledge of the relevant literature.

The two debates that have received most attention are that on the usefulness of randomised controlled trials and on the effectiveness of development aid.

In Deaton 2010, he has been very critical of recent trends in applied economics and in particular in the study of development, that have stressed the use of randomised control trials, often seen as a substitute for theory from which one can learn ‘what works’. He starts from important methodological criticisms and with the identification of many issues (ranging from randomisation protocols to the type of sampling used to build the typical randomised control trial). However, his is most important criticism, in my opinion, is the statement that randomised control trials, while useful, cannot be a substitute for theory. We only understand the working of policies and we can only extrapolate the empirical evidence we derive from data using a well-specified theoretical framework that is brought to bear in a coherent fashion to a set of data. Some of these data can be derived from carefully designed randomised control trials. Others can be well-structured representative surveys, possibly enriched with novel measurement tools aimed at assessing the quantities that are relevant for theory and for our understanding of reality and the way it works.

In the case of international aid, Deaton has become a stern critic of those development economists that advocate an a-critical expansion of aid, ignoring the impact (or lack thereof) that it has. In the final chapters of his recent book The Great Escape, he stresses that while the developed world has a moral obligation towards the billions of individuals that have not escaped poverty and misery in many parts of the world, it cannot ignore that mindless aid funds can do more harm than good, undermining weak states, increasing corruption and possibly preventing real economic growth and development from taking off.

Conclusion

The choice of the Nobel committee in 2015 has been an outstanding one. It has recognised the work of an extremely influential applied economist who has contributed to and combined – in the most profitable fashion – measurement, theory and data analysis to advance our knowledge of consumer behaviour, the determinants of wellbeing and economic development. In the field of consumption, Deaton is one of the few economists who deeply understands the theory of intratemporal and intertemporal allocation of resources and how the theory can talk to the data. He has developed the existing models and made them capable of talking to data. In the process he has influenced generations of applied economists.

References

Browning M, A Deaton, M Irish (1985), “A Profitable Approach to Labor Supply and Commodity Demands over the Life-Cycle”, Econometrica 53: 503-544.

Deaton, A (1974), “The Analysis of Consumer Demand in the United Kingdom 1900-1970”, Econometrica 42(2): 341-367.

Deaton, A (1985), “Panel Data from Time Series of Cross-Sections”, Journal of Econometrics 30(1-2): 109-12.

Deaton, A (1989), “Rice Prices and Income Distribution in Thailand: A Non-Parametric Analysis”, Economic Journal 99(395): 1-37.

Deaton, A (1991), “Saving and Liquidity Constraints”, Econometrica 59(4): 1221-1248.

Deaton, A (2001), “Relative deprivation, inequality, and mortality”, WP No.8099. National Bureau of Economic Research.

Deaton, A (2010), “Instruments, randomization, and learning about development”, Journal of economic literature, 424-455.

Deaton, A (2013), The great escape: health, wealth, and the origins of inequality. Princeton University Press.

Deaton, A and J Y Campbell (1989), “Why Is Consumption so Smooth?”, Review of Economic Studies 56: 357-374.

Deaton, A and J Muellbauer (1980), “An almost ideal demand system”, The American economic review 70(3): 312-326.

Deaton, A and J Muellbauer (1980), Economics and consumer behavior, Cambridge University Press.

Deaton, A and C Paxson (1994), “Intertemporal Choice and Inequality”, Journal of Political Economy 102(3): 437-467.

Deaton, A (1997), The analysis of household surveys: a microeconometric approach to development policy, World Bank Publications.

Deaton, A and C Paxson (1998), “Economies of Scale, Household Size and the Demand for Food”, Journal of Political Economy 106(5): 897-930.

Friedman M, A Theory of the Consumption Function, Princeton University Press, 1957.

Kahneman, D and A Deaton (2010), “High income improves evaluation of life but not emotional well-being”, Proceedings of the national academy of sciences 107(38): 16489-16493.

Modigliani, F and R H Brumberg (1954), “Utility analysis and the consumption function: an interpretation of cross-section data”, in K K Kurihara (ed.), Post-Keynesian Economics, New Brunswick, NJ. Rutgers University Press. Pp. 388–436.

a

A

Topics:  Frontiers of economic research

Tags:  Nobel, consumption, prizes

Professor at University College London

Vox Talks