Disciplinary Note

Bringing Virtues into (Financial) Economics

Susan Thorp

Professor of Finance, The University of Sydney Business School


In the section of her Theology Brief on ‘Virtues Enable Flourishing’, Professor Herdt writes that “The virtuous person does not act well instrumentally… [they] seek to act well because acting well is good, not merely because it is good for me.”

Conventional Economic Models  #

These comments expose a limitation of economics, and financial economics more specifically, where human or corporate behaviour is usually modelled by mathematical models. Typical models comprise two elements: (i) an objective or goal, for example, to satisfy a personal utility such as increased consumption, or to meet production goals in a business firm; and (ii) a constraint, for example, to keep within my personal budget, or to meet a minimum rate of profit. These mathematical models, which are intended to optimize given objectives, are fundamentally instrumental. Everything is about reaching the set objective. Even where the agents (‘me’ or my ‘firm’) of these models extend to groups such as families, corporations, institutions or social planners, all act instrumentally within the structure of the typical economic model. The goals and limitations of the people or firms in the model must be fully described by the components of the model.

Strengths of Economic Models  #

Conventional economic models are useful for description and prediction. They explain incentives, predict a range of behaviours and often expose unintended consequences of regulation and policy, with remarkable clarity. For example, models of increases in taxation of income can show how much some groups of workers might reduce the hours they work when getting a lower after-tax hourly wage, limiting total tax collections.

The instrumental methods of economics help us understand, pragmatically, incentives in a world of both common grace and human sinfulness. These models are capable of capturing self-interest (utility) and altruism (utility that include the interests of others) without making prior judgments about the morality (or virtue) of the objectives.

Limits of Economic Models as Normative Guides  #

Although less conventionally constrained objectives, such as ‘altruism’ (where I gain utility by making charitable donations), social cohesion (where I value how others in a group feel) or ESG goals (where a firm cares about pollution or slavery), marginally change objective functions in these models, they nevertheless fail to grasp the transcendence of virtue on humans and their communities. Thus, these very features of economics mean its models fail as normative guides.

Take the principle of maximising shareholder value, a lynchpin of corporate finance. According to this principle, the prime goal of managers in a firm is to increase the value of shares held by investors. This theory proposes a strict set of unrealistic assumptions that markets will efficiently allocate resources, and individuals and firms do not influence prices, so that the interests of shareholders and society in general coincide. In other words, what is good for the shareholders is good for the society.

However, when used as a normative goal, without regard to market failures, let alone personal and institutional virtues, the outcomes of these models of maximizing shareholder value are predictable, yet sometimes destructive. It is true that the reduction of complex problems in varied social contexts to an instrumental prediction is convenient for teachers (and students) of ‘Finance 101’. It is also the case that this model conveniently offers an ‘escape clause’ for corporations’ choices that neglect virtue and damage people, exploit the environment and degrade community and institutions. Even in recent decades we have witnessed the sub-prime crisis, the oxycodone epidemic, and the misconduct recorded by the Hayne Royal Commission into financial services in Australia. In such cases, shareholder value maximisation was used to justify vicious choices.

As researchers and teachers, we need to state the strengths of disciplinary models alongside their limitations. How are these limitations best presented to our students, colleagues and researchers from other disciplines?

Behavioral Economics  #

At another point, Professor Herdt notes that ‘human beings have both temperamental and acquired dispositions that do not count as virtues’.

Over the past few decades, economic researchers have made radical changes by incorporating personality traits and acquired skill into theory and empirical research. These advances have not been paralleled by similar attention to virtues. For example, in my own field of household finance, conventional approaches now build in human psychological tendencies, such as focusing on the present over the future, and cognitive challenges, such as forecasts involving exponential rather than linear growth (e.g., interest on the interest earned on bank deposits or the spread of infectious diseases).

These psychological traits and skill deficiencies would have been treated as irrelevant 20 years ago. Now they are used to support a variety of policy and regulatory interventions. For example, they influence policy options on automatic or mandatory participation in retirement savings plans in order to combat our bias towards focusing on the present (e.g., Butt et al. 2018). They also shape what are the duties of fiduciaries or advisors who otherwise might be tempted to steer naïve clients towards strategies that lead to higher advisor commissions rather than better outcomes for clients (e.g., Agnew et al. 2018). Behavioural economics and finance have recognized that humans work with informational and cognitive limitations (e.g., Wu et al. 2015), and also that relevant skills can’t be assumed, are variable in the cross-section and over time, and are not always worth acquiring.

Incorporating Virtues in Household Finance  #

While personality traits and skill are increasingly standard features of models, the incorporation of virtue – as a driver of decision-making (e.g., does honesty lead me to return an over payment when there are no apparent consequences or ‘utility’ to not returning it), as a variable personal attribute (e.g., maintaining fair assessment by recognizing that some students will cheat in exams if the opportunity arises, while others will not) and as a dynamic process (e.g., where personal faithfulness is acquired over years of partnership in long-term professional relationships) – seems to have had much less traction. Some studies consider ‘religion’ in limited ways, such as the effect of religious values on readiness to repay debt on time, or the social connectedness gained through church attendance. However, people are not usually defined or modelled by their virtues.

To include virtue, such as mercy, in an optimization model is challenging: if mercy trumps everything, other trade-offs are irrelevant in the decision and the optimization becomes uninteresting. By contrast, some psychological traits can be modelled in trade-offs. For example, the human inclination to value the present over the future (the present bias) can be captured by adjusting mathematical representations of patience and built into otherwise conventional methods. Researchers are more likely to include human characteristics that standard approaches can manage.

Household finance addresses all the financial decisions of life – both frequent day to day decisions like buying groceries, saving for the near and the long term, choosing insurance cover – and large and infrequent choices about education, home loans and retirement. These choices are profoundly influenced by the virtues of honesty (for example, truthfulness on loan applications and tax returns), faithfulness (for example, meeting my loan obligations, working and studying diligently) and self control (for example, staying within my budget).

For social scientists who take virtues seriously, the challenge is to admit the role of virtue and creatively to develop new approaches to a more complete model of behaviour.

Further Reading

Agnew, J. R., Bateman, H., Eckert, C., Iskhakov, F., Louviere, J., & Thorp, S. (2018). First impressions matter: An experimental investigation of online financial advice. Management Science64(1), 288-307.

Butt, A., Donald, M. S., Foster, F. D., Thorp, S., & Warren, G. J. (2018). One size fits all? Tailoring retirement plan defaults. Journal of Economic Behavior & Organization145, 546-566.

Wu, S., Stevens, R., & Thorp, S. (2015). Cohort and target age effects on subjective survival probabilities: Implications for models of the retirement phase. Journal of Economic Dynamics and Control55, 39-56.