Comments

Eduardo Fernández-Arias: This paper is a welcome addition to the literature on the redistribution and poverty effects of economic crises. In addition to the four traditional channels conspiring against the less wealthy (namely, the reduction in labor demand, high inflation, the adverse relative price change associated with real depreciation, and the curtailment of public spending), the authors identify what they term the financial channel. This novel channel appears to consist of financial transfers (or bailouts) in banking sector crises.¹ The paper provides evidence designed to convince the reader that the financial channel is, like the other channels, regressive. It does so by isolating two aspects of bailouts associated with banking crises: the financial transfer from nonparticipants to participants in the banking system (depositors, debtors, and bank owners); and the internal distribution of the transfer among participants. The authors claims that each of these aspects is regressive, and they substantiate these claims with a straightforward accounting of the income profile of payers and beneficiaries of the transfers. Finally, the paper discusses the policy implications of these findings.

In my view, the main contribution of the paper is to raise awareness of this channel and to illustrate how substantial its distributional effects may be. The findings of this paper confirm what most economists (and pretty much everybody else) believe about the regressive characteristics of financial bailouts. These are not minor achievements in a field dominated by the efficiency effects of financial crises and their remedies. In this sense, this is an important paper that opens a research agenda with the potential of altering best-practice policymaking.

However, this paper is only a start, a first pass. It contains three main areas of weakness that future papers ought to strengthen in order to complete the research agenda opened by this paper: (a) the analytical framework used to trace key distributional effects; (b) the quality of the quantitative estimations; and (c) the policy implications discussed. In what follows I elaborate on the analytical pitfalls and then quickly refer to the other two areas of weakness.

Analytical Issues

I highlight three key analytical dimensions that the paper largely overlooks: the economic effects of bailouts; the financing of bailouts; and the ex ante effect of anticipated bailouts.

The Economic Effects of Bailouts

The paper only accounts for the direct effects of bailouts, such as who is entitled to receive the financial bailout in the case of a banking crisis. It thus analyzes the income profile of recipients of deposit insurance schemes to see how depositors benefit; it asks whether debtors benefit from subsidized debt repayment programs, examining the kind of firms that see their debts diluted and the income profile of their owners; and it looks for bank financial support, which would prima facie represent a case of a regressive transfer to wealthy bank owners. As the taxation literature on incidence makes clear, however, these direct or first-round effects may be offset by indirect effects stemming from general equilibrium considerations. Economic modeling and analysis is required to uncover what goes beyond the straightforward, legal incidence taxes, and the same would be true in the case of financial bailouts.

The paper's description and empirical work provides an interesting illustration and is certainly a part of the picture, but redistribution and poverty estimations must deal with some key general equilibrium repercussions that the paper ignores. This weakness can be best appreciated by asking what would happen in the absence of a bailout intended to address a potential or actual banking crisis. The implicit answer of the paper is that the bailout recipients would be worse off (and the payers better off). Nevertheless, many other consequences can be expected, and bailouts exist precisely to alleviate these important consequences. For example, a successful bailout may prevent or contain a systemic banking crisis prompted by lack of confidence. Even within a declared banking crisis, a bailout that eases debt repayment or credit conditions would have a first-order employment effect on the firms kept...