By Robert I. Lerman
Economists often employ cost-benefit methods to analyze government interventions in the economy. Cost-benefit analyses focus on real output: How much real output is used up in a project (social costs) compared with how much real output is generated by the project (social benefits). The Office of Management and Budget has long required cost-benefit analysis in determining whether to undertake a new regulation. The Army Corps of Engineers defends decisions on flood control projects partly based on cost-benefit analysis. The US Department of Labor assesses the effects of its training programs based on experimentally-based estimates of social and private benefits and social costs. Yet, despite the widespread application of cost-benefit analysis to a range of public actions, economic analyses of the tax bill have largely ignored the principles of cost-benefit approaches in favor of focusing almost entirely on government deficits and distributional effects. This is a mistake.
In any comparison, one must take account of timing differences between social costs and social benefits. After all, a dollar in 5 years is worth less than a dollar today because today's dollar could have been invested with a yield of more than a dollar 5 years from now. For example, if you invested a dollar today at a 2.5 percent annual interest rate, today's dollar would be worth $1.13 in 5 years. Putting it another way, today's (present) value of obtaining $1.13 in 5 years is a dollar. The common method for adjusting for timing differences is to calculate the present value of the flows of costs and benefits in various time periods.
How would a cost-benefit approach be applied to the Tax Cuts and Jobs Act? One straightforward social benefit is the increase in real resources resulting from the tax bill, measured as the increase in GDP. The appropriate social cost measure is less clear. One possibility is to view the reduced revenue associated with the new law as a social cost, since the government could have spent the lost dollars to pay for the use of real resources valued by the public. However, the reduced revenue is offset by increased after-tax income that could also have financed the use of real resources enjoyed by the public.
Thus, whether the bill's gross costs represent social costs or reduced transfers from the public to the government (or one set of people to another) depends on how one values the shift in purchasing power. If the costs in terms of real output are minimal, then the bill's projected gains in output are almost certainly higher than the loss in output induced by the tax bill.
Usually, the real resource costs and benefits represent the primary way economists judge whether a government action is a good investment. Actions that use up more resources than they generate leave less to go around, implying it is probably feasible to achieve redistribution more efficiently without the actions. For example, if Job Corps uses up more real resources than the program generates, the government could have saved money and resources by simply giving enough money directly to Job Corps recipients to offset any gains in income.
In the case of the new tax law, the distributional consequences of the bill are complicated and difficult to predict. First, there is the question of who gains from the added value of GDP. A second question is the incidence of the added debt versus the distribution of the tax reductions. Since the bottom half of the income distribution pays practically no net federal individual income taxes, but will receive a significant benefit from the tax bill, high-income taxpayers will presumably owe more in future taxes than they earn from the direct tax reductions.
An infrastructure example can make the analysis more concrete (no pun intended). Let's suppose the government chooses to spend $1 trillion today on new infrastructure projects. In this case, it would use up $1 trillion in resources, ideally in return for future additions to production. We would have to take account of the real administrative and incentive costs of the taxation required to finance $1 trillion today or $1 trillion in the future. If the present value of these increases in GDP exceeded $1 trillion-plus indirect costs, we would conclude that the social benefits exceed the social costs. In addition, we could examine the distributional effects of financing the taxes or debt and of the future GDP gains.
How does the $1 trillion tax cut in the new law differ from an infrastructure project from a cost-benefit perspective? Projected future social benefits may be higher or lower than the infrastructure projects. But the main conceptual difference is that, assuming borrowing is the means of financing the tax cut, the tax bill's costs to the government do not represent a real cost to society, but rather a transfer from future taxpayers to current taxpayers. The $1 trillion in lost current revenues do not use up current resources; from today's perspective, they are offset by $1 trillion in increased after-tax current income along with an increase in the debt obligations of future taxpayers. Note that if an actual infrastructure project were to be financed by debt, it would generate similar obligations for future taxpayers. The only real output costs of the law are the added administrative resources used to adjust to the new law.
Finally, let's suppose we counted the $1 trillion tax bill as a social cost based on the notion that the government could have bought goods and services worth $1 trillion, while household spending from the tax reductions is worth zero. Even in this case, the benefits exceed the cost. Using the Joint Committee on Taxation's growth and revenue estimates, the present value of added GDP over ten years (discounted at 2.5%) is about $1.43 trillion, while the present value of the lost revenue is about $1.32 trillion.
So far, analysts have focused on the likelihood that the tax law's stimulus to GDP will not yield enough tax revenue to offset the loss in tax revenues and will thereby increase government debt. Yet, the law can be justified on cost-benefit grounds, because the substantial benefits in terms of increases in GDP (as projected by credible sources) come with minimal social costs.