Monday, February 11, 2013

Overview of Economics and Psychology

These are originally speaker notes for an introductory talk at the Trinity Economic Forum on Saturday 9th February 2013, but they serve as an effective overview for the MSc Module "Behavioural Economic: Concepts and Theories" I am teaching in Stirling University in the 2013-14 period. You can see the follow up lectures on the widget in the top-left of the blog.

The world is complex and our attention is limited and often distracted so that others can manipulate us. We see the world through a glass darkly. Look at the nobleman in front gazing at the conjurer in Bosch's famous print. Now look at the conjurer's partner behind him stealing his money purse. Much of economic behaviour has this feature of our perception and beliefs leading us into jeopardy, with the thief often being ourselves.

Fig 1. The Conjurer

Fig 2. Adam Smith, father of economics
This is a very interesting time to be Economics and Psychology students. These two expansive disciplines are hitting off one another in ways that are creating fascinating new ideas that are changing how we think about human interactions and having immediate effects on how all of us live our lives in multiple, often hidden, ways. The history of psychology can be traced to scholars in Germany and the US in the 19th century who began to directly investigate mental processes in an empirical manner not hitherto formalized in philosophy. Economics, as a discipline, had no real separate identity in antiquity but finds various expressions throughout the medieval period. While it is almost disrespectful to set the date of its founding, a t-shirt I once saw that read "Economists: confusing people since 1776", of course, bears out the fact that after Adam Smith's epic tome "The Wealth of Nations", it was no longer possible to be considered in the running to be the founder of economics as a discipline.

History of Economics and Psychology 
Fig 3. Gustav Fechner
Given that both disciplines regularly lay claim to be the science understanding human behaviour, you would have thought the interactions would be regular and intensive. And in some sense, they have been. The late 1800s saw widespread attempts to ground Economics in real empirical evidence about the causes of pleasure and pain. The question of utility was examined in many settings and a number of major thinkers including Fechner attempted to place the study of human behaviour on an empirical footing. Marshall famously referred to these attempts at Hedonomics, a phrase he used as a mild put-down.

The portrait of an Economist as a young man
So the question arises as to why you, mostly Economics students, study Economics the way you do and why the Psychology students study Psychology the way they do. I have some advantage in answering this question. I began studying both Psychology and Economics in my local library as a teenager trying to understand the effect the type of massive societal transformations going on around me were having on people. I studied both psychology and economics in Trinity and you will even find one of my first attempts to summarise the link between the two in the Student Economic Review.

Fig 4. Ireland's Net Emigration Rate

Behaviourism in 20th Century Psychology
Fig 5. John Watson
Two major events in US academia explain a large part of the disjunction between the two fields we see at undergraduate level very starkly. On the one hand, Psychology began to strive more for empirical realism, culminating in the dominance of behaviourism as an overarching paradigm for the field. Armed with an array of experimental paradigms and the development of a set of laws of association and conditioning, the behaviourists, led by John Watson, set out to explain all human behaviour as the product of associations built up over life through conditioning.

Economics as Axiomatic Science 
Fig 6. Paul Samuelson
Meanwhile, economists were beginning to suffer physics envy, with many of the leading US academics seeking to set Economics on a sound axiomatic footing. The publication of major works such as Samuelson's formulation of discounted utility theory, the development of Arrow-Debreu equilibrium theory, the development of game theory and many related intellectual advancements propelled Economics and rational-choice Economics, in particular, into the forefront of Western thinking, guiding thought about how markets should be organised and regulated even questions such as how cold-war strategy should be played.

The Economics we teach 
And so it came to pass that Economics textbooks were increasingly populated with axiomatic proofs built up from elementary concepts of human behaviour, using notions of optimality and equilibria that could be applied to a range of economic problems. You study Economics very much as a discipline where rational, utility and profit-maximising individuals and firms strategically interact under different market structure constraints. Such models then took on a life of their own. Feedback loops between Economics and the real-world led to many of these models being thought about and increasingly taught as if they were naturally occurring phenomenon and not models that theorists had invented to solve particular problems at particular times.

Why Economics textbooks tell only part of the story 
Fig 7. John Maynard Keynes
And, even worse, all of this is a very shallow description of what many of these people actually wrote. Samuelson never believed that individuals performed the type of calculations required in his discounted utility theory. Reading the textbook IS-LM model would make one believe that Keynes was a fairly mild figure with a constrained view of how humans interacted. And yet reading Keynes directly, and you are struck with how intensely he thought about the psychology of investment and consumption. The General Theory is as much what we would call behavioural economics today as any modern work. And, shock horror, even Adam Smith's other major work "The Theory of Moral Sentiments" is a lengthy meditation on how people behave in social situations, with whole chapters on altruism, morality, convention and all of those other things we assume away quite early in an Economics education.

The cognitive revolution collides back with Economics 
Fig 8. Daniel Kahneman
Most excitingly, the 60s and 70s saw the emergence of thinkers directly challenging the dominance of rational choice models in Economics. The seminal work of Herbert Simon opened up the formal study of bounded rationality that has had implications across science.
The psychologists Daniel Kahneman (who would later win the Economics Nobel in 2002) and Amos Tversky also did seminal work during this period, most famously with their "Prospect Theory" paper, which acted as a descriptive model of peoples behaviour in contrast to the normative approach of Expected Utility.

Rethinking our assumptions 
Thinking through these issues raises profound questions. I urge you to look back at the assumptions made in Economics and use them as rocket ships to explore the vast universe of economics and policy.

1. Preferences are complete.

2. People discount the future exponentially.

3. People make rational judgements with the information available.

4. People prefer more choice to less in all circumstances.

5. People seek to maximise utility.

6. People value losses and gains symmetrically.

7. People care for their own/family utility only.

But, there are so many challenges to these assumptions:

1. Can we ever say preferences are complete? In some market, this may be more realistic than others but the set of options is a variable and never fully known.

2. Exponential discounting increasingly looks a poor account of how people process the future.

3. The extent to which people prefer more options may also be a variable dependent on interactions between their own characteristics and the choice envrionment.

4. Loss aversion is heavily observed both in humans and animals and places important constraints on markets and trade.

5. Identity considerations can lead to patterns of behaviour that look completely at variance with an attempt to live a long, healthy and wealthy life.

6. Depending on context, we may also value others utility to a higher degree than suggested in the standard model.

Pensions and the Life-Cycle model of consumption 
One of the best examples of this area is pension policy. In 2014 Ireland is scheduled to become one of the first countries to move toward a national automatic enrolment pension system for the private sector. All non-covered private sector workers will be automatically enrolled into a default privately provided pension scheme, further incentivised by mandatory employer co-contributions and tax incentives.

Benartzi and Thaler (2004) is one of the most cited examples of intervention in private pension provision. Save More Tomorrow (SMarT) is a savings plan that attempts to overcome behavioural biases in saving for retirement; these biases include: hyperbolic discounting, loss aversion, and inertia. The perceived need for intervention in private pensions arises from a shift to defined-contribution (DC) pensions, with DC pension plans supplanting defined-benefit (DB) schemes, with the onus increasingly being on employees to provide for retirement.

The authors cite four principles on which their plan is based:

1) An employee should be approached as early as possible before a scheduled pay increase, with a commitment to save more. This exploits hyperbolic discounting in favour of the plan, where the ‘loss’ is not immediate, but will occur in the future.

2) The increased contributions should take place immediately after a pay increase to mitigate a ‘loss aversion' effect.

3) For each scheduled pay increase, the contribution rates rises until it reaches a specified maximum. This utilises employee inertia in favour of the savings plan.

4) An employee can opt-out of the plan at any time.

Three firms used the SMarT programme, with some heterogeneity in implementation. The first (and most comprehensive) implementation was at a midsize manufacturing firm, where employees on lower income were not saving sufficiently in the view of management of the firm. This dearth of saving created a problem for executives: they could not contribute the maximum tax allowable amount to their pension plans due to U.S. Dept. of Labor non-discrimination laws—non-discrimination laws try to prevent the tax benefits of pension savings schemes accruing disproportionately to higher paid employees, defined as those earning above $110,000 per annum in 2010.

Employees were given a time slot to discuss their retirement plan with a financial consultant (FC in Fig 9 below). Out of 315 eligible employees, only 29 chose not to meet the FC. Employees were offered a savings plan based on their conversation with the financial consultant, where a maximum of 5 per cent savings rate was proposed for those who stated financial difficulty in increasing their contribution, or a rate specified by commercial software if the employee was in a better position and appeared willing to increase their savings rate beyond 5 per cent,  in which case the amount given was usually the tax-allowable limit. The SMarT plan was offered to those who rejected the financial consultant’s advice, which carried a rise of 3 per cent in savings rate with each future pay rise.

Figure 9 below (derived from figures in the paper) summarizes the outcome of different choices by workers in the first company before and after pay rises; the first bar cluster represents savings rates prior to employees meeting with the financial advisor, with subsequent clusters being savings rates between the first and fourth pay raises. The jump from 3.5 per cent to 13.6 per cent savings rate for those who entered the SMarT plan is substantial.  Furthermore, 80 per cent of SMarT plan participants stayed on to the fourth pay rise.

Fig 9. Save More Tomorrow

One seminal paper on automatic enrollment is Madrian and Shea (2001). The authors analyse a firm that changed its pension enrollment criteria from employees choosing to opt-in after one year in the firm, to a scheme where, upon being employed by the firm, employees are automatically enrolled. The reason for changing their enrollment policy was due to the firm continually failing non-discrimination tests, and thus needing to make costly ex-post refunds to ‘highly compensated’ employees.

Figure 10 below shows the increase in participation by Ethnicity and Race, with the left block representing the cohort of employees who took the decision to enroll in the firm’s pension plan upon becoming eligible, with 3 to 15 months of tenure, and the right block represents those employees who were automatically enrolled, again with 3 to 15 months of tenure. The most dramatic effect is on the Black and Hispanic workers, with enrollment nearly quadrupling for these groups.

Fig 10. Madrian & Shea's automatic enrollment
Identity and Motivation 
Fig 11. Identity Economics
The key paper for this is Economics and Identity by Akerlof and Kranton. This paper takes the view that looking at identity is vital to understand a wide range of economic phenomenon such as welfare dependency, ghettos, integration into the labour market, globalization and economic growth.

Identity emerges from the social categories we identify with or are members of by default. They outline a very simple model, which we will cover in the lecture, where membership of social categories enters directly into utility functions and use this to explain a range of economic phenomena such as gender discrimination.

Politics and Policy 
Fig 12. Nudge
The political implications of this type of policy has not gone unnoticed. In the UK, the Conservatives have brought behavioural economics right to the heart of government with the development of the behavioural insights team. And it does raise very fundamental questions about the relationship between individuals, large corporations and government. If people make mistakes in decisions (e.g. by being confused, misjudging risk etc.,) then their decisions may no longer be a good guide to their actual welfare. This would imply that the market system itself may not necessarily yield the best outcomes for consumers. It also potentially implies that giving consumers more information, framing that information them more clearly etc., might improve their welfare. Thus the main public policy implication would be that it might be possible to improve people's welfare through some of these mechanisms, something that would be explicitly ruled out by a model where people fully rationally interpret risk.

People like Thaler and Sunstein have used the phrase “Libertarian Paternalism” to get at the idea that it is possible to shape policy such to make it easier for people to make optimal choices from their own perspective (e.g. taking out a pension) without forcing them to do. The pension opt-outs are an example of this type of policy though in the Irish case, it is somewhat more forceful as you are opted back in every two years, a type of Catholic church style nudge.

Economics and intellectual exploration 
In conclusion, Economics is an intense questioning of how people behave and how we can design institutions. Each of the sometimes dry assumptions we make in microeconomic textbooks emerged from centuries of debate. They are living and moving assumptions. Each one potentially unlocks the key to new institutions and new ways of living. I wont, like Robin Williams, ask you to tear up your textbooks. They are too expensive. But do tear up your assumptions. Participate in this debate.

1. Carroll et al. (2009), Optimal Defaults and Active Decisions, Quarterly Journal of Economics.
2. Benartzi & Thaler, How Much is Investor Autonomy Worth?The Journal of Finance.
3. Benartzi & Thaler, Save More Tomorrow: Using BehavioralEconomics to Increase Employee Saving, Journal of Political Economy.
4. Choi et al. (2006), Reducing the Complexity Costs of 401(k) Participation Through Quick Enrollment, NBER Working Paper.
5. Madrian & Shea, The Power of Suggestion: Inertia in 401(k) Participation and Savings Behavior, Quarterly Journal of Economics.
6. Sunstein & Thaler (2003), Libertarian Paternalism, American Economic Review.

Popular texts:
1. Ariely (2008), Predictably irrational.
2. Kahneman (2011), Thinking, fast and slow.
3. Thaler & Sunstein (2008), Nudge: Improving Decisions About Health, Wealth, and Happiness

1. Camerer, Loewenstein & Rabin (2004), Advances in Behavioral Economics
4. Shafir (2013), The Behavioral Foundations of Public Policy

Policy-related documents:
1. Dolan et al. (2012), Influencing behaviour: the mindspace way, Journal of Economic Psychology.
3. House of Lords Science and Technology Select Committee (2011), Behaviour Change Report, London: TSO.

No comments: