Why this study deserves our attention

Summary

“Studying key events related to the repeal of Dodd-Frank policies we find that banking deregulation can create shareholder value. We document positive stock returns for banks around these events. The returns are larger for small banks with assets below $10 billion suggesting that their shareholders expect greater benefits. Furthermore, we observe improvements in capital strength and accounting performance, an expansion in external financing and lending activity, and positive asset growth for only small banks. The eased regulation for large banks with assets above $50 billion likely made small banks more willing to grow again and cross the $10 billion threshold.

Overview

  1. What is regulation for? How to interpret the positive market response to deregulation?
  2. When you have a hammer everything looks like a nail. Is it possible, in this setting, to create a set of meaningful “normal” returns?
  3. Does size or number matter? The effects are opposite for the majority of banks (banks <10bn) and the banks that hold the majority of the assets (banks

    10bn):

    • So what is the effect?
    • Why is deregulation good for small banks, but not large banks, who are actually being deregulated?

1. What is regulation for?

The authors interpret the positive market reaction to deregulation as evidence in favor of deregulation.

1. What is regulation for?

From the introduction:

“Our event study yields two major findings that lend support to the pro-deregulation arguments: First we show a mean increase of 1.41 percent in banks’ three-day abnormal returns around the announcement of deregulatory events.1

From the Columbia Blue Sky Blog:

“These positive outcomes are consistent with the stronger stock market reaction revealed in our event study and lend credibility to shareholders’ positive assessment of financial market deregulation.””

1. What is regulation for?

1. What is regulation for?

  1. Arguments based on shareholder value alone assume that the market can internalize the external costs or solve the agency problem that the regulation was created to manage.
  2. We would expect a positive market reaction around the relaxation of a regulation that imposed a cost on the regulated firms, whether it was effective or not.

Alone, market-based tests are not capable of providing evidence in favor of deregulation.

2. When you have a hammer everything looks like a nail

2. When you have a hammer everything looks like a nail

This is uniquely problematic, because in this study all banks experience the event at the same time. In contrast to something like mergers, where only certain banks are impacted so similar banks are available for comparison.

2. When you have a hammer everything looks like a nail

I think that the authors have thought about this, because they write:

“To alleviate potential bias in our inference, we screen for confounding events that took place within 11 days surrounding each event to check that these events do not fit any discernible pattern.” (p.16)

It is worth thinking about whether potential confounding events have differential effects on banks and non-banks. For example, only a few of these fall are more than 11 days away from Fed meetings. So it is hard to argue that non-banks ever isolate a unique deregulation shock from general bank shocks.

Caveat: People do this

Here are two papers that use a relatively similar approach to the one taken in the paper:

“Asymmetric information, dividend reductions, and contagion effects in bank stock returns,” (Bessler and Nohel 2000, JBF)

“Common stock returns in corporate takeover bids: Evidence from interstate bank mergers,” (Cornett and De 1991, JBF)

2. Illustrations

3. Does number or size matter?

Table 4 Col 1 shows a positive treatment effect for small relative to large banks, and a negative treatment effect for large banks (the ones that are actually deregulated). This pattern appears throughout the results.

Two questions

3. Does number and size matter?

Additional Items

It’s unclear how important the $10 bn and $50 bn cuts are to the paper

Are these banks?

“All 25 events are connected to the regulatory landscape of firms in the traditional banking industry. Thus, our sample includes only U.S. traditional banks with the necessary price and financial statement data from 2016 to 2018. We start with 534 domestic U.S. banks (two-digit SIC codes 60, 61, and 62).”

Group names:

What is a bank?

(1) In general.—Except as provided in paragraph (2), the term “bank” means any of the following: Bank Holding Company Act of 1956 (12 U.S. Code § 1841): (A) An insured bank as defined in section 3(h) of the Federal Deposit Insurance Act [12 U.S.C. 1813(h)]. (B) “An institution organized under the laws of the United States, any State of the United States, the District of Columbia, any territory of the United States, Puerto Rico, Guam, American Samoa, or the Virgin Islands which both (i) accepts demand deposits or deposits that the depositor may withdraw by check or similar means for payment to third parties or others; and (ii) is engaged in the business of making commercial loans.”

  1. Potentially deregulatory events? These are not events of deregulation, but steps in the deregulatory process. 

  2. Or for banks that are being impacted less, if not completely un-impacted. 

  3. The way that the subcolumns are set up in table 4 makes the relation nearly mechanical.