Tax specialists normally don’t focus on financial stability and financial regulators and analysts typically do not focus on taxes. This is too bad because the corporate tax structure affects financial stability and does so significantly, as we analyze in this article.

The reason is simple: tax rules influence the capital structure choices of corporations in general and banks in particular, by allowing tax benefits to debt—principally the deductibility of interest—that equity lacks. Today, the corporate tax penalizes equity finance and subsidizes debt relative to equity. As a consequence, corporations overall use more debt than they would in a tax-neutral world.

The problem is particularly pernicious for banks, whose equity levels tend to be naturally thin. The tax penalty for equity and the subsidy for debt, as we demonstrate in this paper, is a key but underappreciated reason for banks’ recurring excessive risk taking and weaker-than-optimal capitalization.

A bank’s raw material is not steel or electricity, but funding; and the tax cost of its funding affects the bank’s choice of capital structure and its overall cost of funding. A conservatively-financed bank with a high level of equity will pay a higher tax load for every dollar of loans it provides. The bank will have to quote higher interest rates to its customers and eventually will lose market share to more levered and riskier but less taxed financial channels.

To avoid this problem banks often seek to undermine regulators’ capital adequacy rules, either transactionally or by lobbying for their repeal. The resulting debt-heavy structure not only renders banks fragile but also pushes them toward excessively risky strategies.

This corporate tax distortion is well known among tax analysts, but it is not inevitable. The recent tax reform, by lowering the corporate tax rate reduces the tax disparity between debt and equity. It is, however, entirely possible to structure taxes in a way that fully eliminates the disparity — and which would have a noticeable impact on bank equity levels, making banks noticeably more stable. And it is even possible to structure bank taxation to favor instead of punishing banks with high equity levels.

In this paper we show how this can be done in a revenue-neutral way—so that the overall level of tax revenue that the government raises from the financial sector stays steady. Several means to the desired end are possible. We discuss each, but the best trade-off between administrability and effectiveness is a system that gives banks a tax allowance for non-regulatory equity, according high-equity banks a deduction for the cost of equity similar to the deduction that they would have received if they had used debt instead of equity. This solution — treating some of the cost of equity parallel to the tax treatment of the cost of debt — would minimize the impact on government revenues (because regulatory equity would be taxed as it is now) and would maximize the banks incentives to hold strong equity buffers above the regulatory minimum. Any loss of tax revenues would be small (primarily because the plan would substitute newly tax-favored equity for currently tax-favored debt) and could be offset by modest tax changes targeted at the riskiest forms of financial sector debt.

With a pro-equity bias instead of a pro debt bias, banks’ tax incentives would be aligned with the regulator’s incentives and the public’s. Several countries have implemented tax laws that are a step in this direction and the evidence from these countries demonstrates than banks would voluntarily increase their equity levels to save taxes. We analyze the approximately two dozen post-2010 studies of the impact on equity levels of corporate tax rate changes on banks in particular and corporations generally.

With that analysis we estimate the likely impact of our proposal to change the way bank equity is taxed. The impact is likely to be substantial. Indeed, the magnitude of the resulting safety benefit should rival the size of the benefit from all the post-crisis capital regulation to date. Thus the main thesis we bring forward is not a small or technical claim.

Standard bank regulatory style is command-and-control, and while much can be and has been accomplished with the standard style, it has its limits. In today’s political environment, current safety rules’ continuance may not be viable, as a repeal of recent regulatory advances, rather than refinement, has become a serious possibility. Yet rolling back the post-crisis regulatory advances without addressing the underlying risk-taking incentives would be unwise. While our policy preference would be to supplement and not replace traditional and recent regulation with the tax reform, any major rollback makes reducing the risk-taking tax distortion more urgent than ever.

Mark Roe is the David Berg Professor of Law at Harvard Law School. Michael Troege is a Professor at ESCP Europe.

This post is adapted from an article published in Volume 35, Issue 1 of the Yale Journal on Regulation.