Economic Theory

Can Behavioural Economics improve economic policymaking?


Strategies based on Behavioural Economics — ‘nudges’— now inform policy thinking in areas as diverse as obesity, tax compliance, smoking, energy use, consumer protection, drug abuse, retirement savings, and organ donations, to name just a few.

When it comes to voting, nudge proposals have mostly been restricted to increasing voter turnout. But I think we should go beyond this, to consider using some of the insights from Behavioural Economics to improve electoral outcomes. Here are some ideas to provoke debate.

My first three proposals deal with fiscal matters. Many of us might disparage voters’ preferences for an extensive, costly government. But the question we need to ask is: are these genuine voter preferences or a mere artefact of the payment method we use?

The first proposal is the elimination of tax withholding (in the UK, the PAYE system). This is not an original suggestion. But the rise of Behavioural Economics has provided a very robust foundation for it that is worth reconstructing. In its essence, tax withholding reduces the perceived costs of government policies by precluding voters’ development of their psychological sense of entitlement over the withheld portion of their paychecks. Thus, the ‘loss aversion bias’ is nullified. It is not a coincidence that Council Tax is one of the most unpopular taxes in Britain, even though it only accounts for a small part of the total tax bill. Council Tax is one of the few taxes that are not withheld; rather, we have to pay it out of our bank accounts, and thus, from money we already have, and perceive as “ours”.

My second proposal is the distribution of itemised tax bills (as opposed to tax receipts or statements). Itemised tax bills requesting immediate action would capture the limited time and attention of taxpayers, as well as bypass their tendencies to procrastination. These bills might also enable taxpayers to arrive at a more informed understanding of their own policy priorities; allowing them to recognise the opportunity cost of each policy both in terms of other policies that they would like to see implemented and their own private consumption and saving.

There is little doubt that politicians prefer to rely on debt over taxes because of the lower salience of the costs imposed over their voters. To rectify this, my third proposal consists in distributing a notice of public debt to taxpayers. However, for the same considerations offered in support of tax bills, the notice of public debt should also serve as a bill for the interest paid on it. Separating the payments of the interest on public debt from that of the other budget expenditure categories would reduce politicians’ incentives to rely on deficit spending to lower the perceived costs of government. It would also contribute to increasing voters’ understanding of the allocation of responsibilities between current and past administrations.

In markets where providers have no incentives to correct consumers’ biases and provide full information (such as those for savings plans) disclosure mandates are one of the common policies to protect consumers and minimise negative social consequences. Itemised tax bills and notices of debt would work as disclosures when it comes to the total cost and distribution of government spending.

However, not all costs imposed by government are charged to taxpayers in the form of taxes and debt. So it is worth considering the potential benefits of an annual policy disclosure. In the United States, the president’s budget proposal, typically submitted during the first week of February, contains detailed information on spending and revenue proposals for the fiscal year starting on 1 October, along with policy proposals and initiatives with significant budgetary implications. In March, the non-partisan Congressional Budget Office publishes an analysis of the president’s budget proposals. This analysis could constitute the main source for the substantive information to be included in the annual policy disclosure for voters. It might also include estimates of the private costs of, for example, new employment mandates.

By themselves, policy disclosures would probably have limited effects among the majority of voters because of their lack of interest, limited attention, and the information’s potential to upset their beliefs and affective commitments. So how about also considering the merits of a final proposal: a refundable tax credit for political competence to be granted on the basis of some sort of testing of individuals’ familiarity with the information contained in policy disclosures.

Such tax credits would provide a private benefit that is linked to an individual’s investment in reducing political ignorance but that is independent of collective electoral outcomes over which the individual has no control. Tax credits for political competence could also help in reinforcing the currently weak social norm regarding the unacceptability of biased and uninformed voting.

These proposals should be expected to raise a host of questions and challenges worth exploring. In a paper in the June issue of the IEA’s journal, Economic Affairs, I attempt to address some of the most pressing ones.



Newsletter Signup