Plan of the Study and Principal Findings
A core argument in the Admati-Hellwig view is that financial theory posits that there should be no cost at all from higher equity and less debt leverage because of the Modigliani-Miller (M&M) theorem whereby capital structure is “irrelevant” (Modigliani and Miller 1958).
The proposition is that equity investors will be satisfied with a lower rate of return if the firm has more equity and less debt, because the riskiness of the firm will have declined, and that the resulting decline in the unit cost of equity investors demand will be just enough to offset the shift from low-cost debt to higher-cost equity. The first central question in this policy area, then, is whether this theorem is operationally reliable for banks or breaks down in practice.Chapter 2 provides a critical review of the literature on capital requirements. Chapter 3 begins the quantitative analysis with an empirical test of the M&M theorem for US banks. With an estimate in hand for the effective degree of the M&M “offset” (extent to which higher quantity of equity is offset by lower unit cost), chapter 4 turns to quantification of benefits and costs of higher bank capital and identifies the optimal capital ratio. Chapter 5 addresses TLAC requirements, both surveying the literature and providing new evidence on the losses of the largest US banks in the Great Recession, to determine whether they had engaged in excessive risk taking as a result of too big to fail (TBTF) incentive distortions. Chapter 6 critiques the recent empirical literature maintaining that there is already too much finance, based on statistical correlations of growth with indicators of financial depth. If it were the case that finance is excessive, the curbing of bank activity by higher capital requirements might bring a supplementary benefit by shrinking finance back to the optimal size, but the tests in chapter 6 do not support this implicit proposition. Chapter 7 summarizes the study's main findings.
The principal quantitative findings of this study are as follows:
■ Only about 45 percent of the M&M risk offset attains in practice (chapter 3).
■ Each percentage point increase in capital required relative to total assets (not risk-weighted assets) reduces the long-run level of GDP by 0.15 percentage points (chapter 4).
■ A banking crisis imposes damage equal to about two-thirds of one year's GDP in present-value terms, including longer-term effects (chapter 4).
■ The optimal level of tangible common equity is 7 to 8 percent of total assets, corresponding to 12 to 14 percent of risk-weighted assets (chapter 4).
■ Correspondingly, the optimal capital ratio is about one-third larger than the target set in Basel III (13 percent of risk-weighted assets compared with 9.5 percent for tangible common equity for G-SIBs). In practice, however, US banks hold close to the optimal amount, because they maintain a cushion above the required minimum (chapter 4).
■ In the Great Recession, the largest US banks did not experience losses that were proportionally significantly greater than those of mediumsize banks, casting doubt on the proposition of excessive risk taking spurred by TBTF incentive distortions (chapter 5).
■ New tests cast doubt on the too much finance literature (chapter 6).