PrefBlog

This paper was referenced in the recent BoC analysis of capital ratio cost/benefits as the “LEI Report”.

The Bank for International Settlements has released a report titled An assessment of the long-term economic
impact of stronger capital and liquidity requirements:

This simple mapping yields two key results, with the central tendency across countries measured by the median estimate. First, each 1 percentage point increase in the capital ratio raises loan spreads by 13 basis points. Second, the additional cost of meeting the liquidity standard amounts to around 25 basis points in lending spreads when risk-weighted assets (RWA) are left unchanged; however, it drops to 14 basis points or less after taking account of the fall in RWA and the corresponding lower regulatory capital needs associated with the higher holdings of low-risk assets.

Their approach relies heavily on the theory that output losses due to bank crises may be ascribed entirely to the crisis (although they do acknowledge that “that factors unrelated to banking crises, and not well controlled for in these studies, may also influence the output losses observed in the data.”):

Why should the effects of banking crises be so long-lasting, and possibly even permanent? One reason is that banking crises intensify the depth of recessions, leaving deeper scars than typical recessions. Possible reasons for why banking related crises are deeper include: a collapse in confidence; an increase in risk aversion; disruptions in financial intermediation (credit crunch, misallocation of credit); indirect effects associated with the impact on fiscal policy (increase in public sector debt and taxation); or a permanent loss of human capital during the slump (traditional hysteresis effects).

One of the papers they quote in support of their approach is Carlos D. Ramirez, Bank Fragility, ‘Money Under the Mattress,’ and Long-Run Growth: U.S. Evidence from the ‘Perfect’ Panic of 1893:

This paper examines how the U.S. financial crisis of 1893 affected state output growth between 1900 and 1930. The results indicate that a 1% increase in bank instability reduces output growth by about 5%. A comparison of the cases of Nebraska, with one of the highest bank failure rates, and West Virginia, which did not experience a single bank failure reveals that disintermediation affected growth through a portfolio change among savers – people simply stop trusting banks. Time series evidence from newspapers indicate that articles with the words “money hidden” significantly increase after banking crises, and die off slowly over time.

Ramirez continues:

The intuition behind the explanation of why financial disintermediation affects
growth is straightforward. In the absence of deposit insurance or any other institutional arrangement that restores confidence on the banking system, depositors who experience losses or whose money becomes illiquid, even temporarily, may become reluctant to keep their money in the banking system. They simply stop “trusting” banks. This lack of trust may affect all depositors, including those that did not experience losses. With a high enough degree of risk aversion and a high enough probability of a bank run or failure depositors may be induced to reshuffle their liquid asset portfolio away from the banking system. To the extent that the panic induces a portfolio change in asset holdings away from the banking system and into more rudimentary forms of savings, such as keeping the money “under the mattress,” financial intermediation, and thus, growth are adversely affected.

According to me, this raises a red flag about the use of these data to determine the the severity of bank crises in the current environment, even without considering the difficulty of disentangling the degree of recession actually caused by the Panic versus the degree of economic stupidity that was merely brought to light. The Nebraskans kept their money “under the mattress” due to a lack of deposit insurance – just how relevant are the mechanics to what is going on now?

Additionally, the underlying rationale behind the desire to avoid banking crises points to an alternative solution to the problem: rather than reducing the chance of a systemic banking crisis, why not increase the range of alternative intermediation pathways?

Currently, regulators are doing everything they can – by way of Sarbox, completely random regulatory punishment for having been involved in the underwriting and distribution of investments that went bad, TRACE, costs of prospectus preparation, etc. – to deprecate the direct capital markets. A concious effort should be made in the other direction; the option of issuing public debt should be made more attractive to companies, not less.

Additionally, I have previously proposed that Investment Banks be treated differently from Vanilla Banks, not by a strick separation such as Glass-Steagall, but by imposing differing schedules for different types of banks, so that the latter would be penalized for holding assets while the former would be penalized for trading them (where “penalized” refers to higher capital requirements). The idea can be extended: bring back the Trust Company, which would get a preferential capital rate for first mortgages with loan-to-value of less than 75% and a penalty rate of everything else.

In the financial system as in the financial investments, bad investments (bad banks) can hurt you: it is concentration that kills you. And the point is related to my other big complaint about the regulatory response to the crisis: right in the age of networking, regulators are emphasizing systems which are vulnerable to single point failure.

Just as an example, the BIS paper notes:

The final approach used in this exercise relies on the Bank of Canada’s stress testing framework. This methodology is based on the idea that the failure of a bank arises from either a macroeconomic shock or spillover effects from other distressed banks. Spillover effects arise either because of counterparty exposures in the interbank market or because of asset fire sales that affect the mark to market value of banks’ portfolios. In this context, a greater buffer of liquid assets can only be beneficial insofar as it helps the bank to avoid asset fire sales, which would otherwise lead to losses.

You could get the same benefit by reducing the degree of interbank holdings, but such a solution is not even considered. BIS, it appears, will continue to risk-weight bank paper according to the credit rating of the sovereign – if you want an example of moral hazard, that’s a good one right there.

Anyway, BIS comes up with a figure of 13bp per point of capital ratio:

Column A of Table 6 reports the results of this exercise. In order to keep ROE from changing, each percentage point increase in the ratio of TCE to RWA results in a median increase in lending spreads across countries of 13 basis points.

… but this is subject to four major problems acknowledged by BIS:

• The existing literature, which is the basis for this report’s estimates of the costs of banking crises, may overestimate the costs of banking crises. Possible reasons include: overestimation of the underlying growth path prior to the crises; failure to account for the temporarily higher growth during that phase; and failure to fully control for factors other than a banking crises per se that may contribute to output declines during the crisis and beyond, including a failure to accurately reflect causal relationships.
• Capital and liquidity requirements may be less effective in reducing the probability of banking crises than suggested by the approaches used in the study. This would reduce the overall net benefits for a given level of the requirements. However, to the extent that net benefits remain positive, it would also imply that the requirements would need to be raised by more in order to achieve a given net benefit.
• Shifting of risk into the non-regulated sector could reduce the financial stability benefits.
• The results of the impact of regulatory requirements on lending spreads are based on aggregate balance sheets within individual countries, so that they do not consider the incidence of the requirements across institutions. They implicitly assume that theinstitutions that fall short of the requirements (ie, that are constrained) do not react more than those with excess capital or liquidity (ie, that are unconstrained). These effects may not be purely distributional.

I consider the third point most important. To the extent that higher capital ratios imply higher spreads implies a greater role for the shadow banking industry, then the real effect of higher capital levels will be to shift important economic activity from the somewhat flawed regulated sector to a sector with no regulation at all.

I certainly support the idea that we should have layers of regulation – a strong banking system surrounded by a less regulated / less systemically important investment banking sector, surrounded in turn by a wild-n-wooly hedge fund/shadow bank sector … but regulators are, as usual fixing yesterday’s problem with little thought for the implications.