Annals of captured regulators, NY Fed edition

By Felix Salmon
March 20, 2014
Peter Eavis has a worrying story today: the chairman of the New York Fed, William Dudley, has effectively, behind the scenes, managed to delay the implementation of an important new piece of bank regulation.

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Peter Eavis has a worrying story today: the chairman of the New York Fed, William Dudley, has effectively, behind the scenes, managed to delay the implementation of an important new piece of bank regulation.

The first thing to remember here is that delaying regulations is an extremely profitable game for the financial industry. If a new regulation will cost a bank $100 million per year, and the bank gets that new regulation delayed by a year, then it’s just made $100 million in excess profit. What’s more, the further away you get from the crisis, the harder it becomes for new rules to grow teeth. So when the banking lobby doesn’t like a certain piece of regulation, its tool of choice is to bog it down and delay it to the point at which no one but the banking lobby cares any more. And then allow it to be implemented with so many loopholes and carve-outs that it’s effectively toothless.

In this game, the banks are on one side, and the regulators — primarily the Federal Reserve — are on the other. So it’s particularly worrying when a regulator ends up causing a delay and thereby helping the banks. And yet that’s exactly what seems to have happened:

Mr. Dudley’s concerns played a decisive role in holding up the final version of the rule, two of the people said. Some regulators, including officials at the Federal Deposit Insurance Corporation, were counting on the leverage regulation being completed by the end of last year. Strong supporters of the rule wanted it issued by then to reduce the chances that pressure from bank lobbyists would dilute it. The rule is now expected to come out in April at the earliest.

The optics here are not helped by the fact that Dudley made his millions at Goldman Sachs, a bank which would be directly affected by the rule in question, which forces big banks to increase the amount of capital that they hold against their assets. Neither are they helped by the fact that Dudley runs the New York Fed, which is generally seen as the arm of the Fed which is closest to, and friendliest with, America’s biggest banks. (Indeed, JP Morgan’s Jamie Dimon was a member of the board there for the six eventful years to 2013.)

Mostly, however, the problem is that Dudley’s objection is very silly.

Mr. Dudley raised the possibility that the rule could inhibit the Fed’s ability to conduct monetary policy…

The Fed officials in Washington assessed his concerns but did not think they were serious enough to warrant significant changes to the rule, the three people said.

In theory, Dudley is right. The way that the Fed conducts monetary policy is by instructing the traders at the New York Fed to buy and sell certain financial instruments so that a particular interest rate — the Fed funds rate — is very close to a certain target. Through a complex series of financial interlinkages, setting the Fed funds rate at a certain level then has a knock-on effect, and ultimately helps determine every interest rate in America, from the Treasury yield curve to the amount you pay for your credit card or your mortgage.

Those interlinkages are so complex that they’re impossible to model with any particular accuracy: all the Fed can do, really, is set the Fed funds rate and then see what happens to everything else. And directionally the causality is clear: if the Fed wants rates to rise, then it pushes the Fed funds rate upwards, and if it wants rates to fall, then it brings the Fed funds rate down. That doesn’t always work at the distant end of the yield curve, but it’s still most of what monetary policy can do.

Especially early on in the chain, a lot of the interlinkages take place at the level of big banks. And so it stands to reason that if you change the leverage requirements of big banks, that might change what happens to interest rates when you move the Fed funds rate. Or, on the other hand, it might not. In any case, if and when the Fed starts raising the Fed funds rate, it’ll rapidly become pretty obvious what’s happening to the rest of the interest-rate world, and the FOMC will react accordingly.

In the most extreme case, the FOMC might even change the way it sets interest rates, and start using interest rates other than the Fed funds rate to conduct monetary policy. After all, the Fed can intervene pretty much anywhere in the financial system it likes. Obviously, Dudley, as the head of the New York Fed, would be the person most closely consulted in terms of determining the most effective way for the Fed to intervene and move American interest rates. And in making his recommendations, he would have to take into account everything he knows about the architecture of the financial system, including the leverage ratios being demanded of the biggest banks.

But what doesn’t make sense is the idea that Dudley would try to throw a spanner in the works of an important piece of bank regulation, just because it might make his rate-setting job more difficult. The New York Fed is a highly profitable institution which employs a large number of extremely able traders and economists, all of whom are well versed in navigating the complexities of the interest-rate market. If a change to the leverage rule makes their job a bit more interesting or difficult, well, that’s part and parcel of what it means to work at the New York Fed. It’s no reason at all to delay a rule change and give New York’s banks a gift on a plate.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
2 comments so far

Felix

“In this game, the banks are on one side, and the regulators — primarily the Federal Reserve — are on the other”

To call the Federal Reserve a ‘bank regulator’ at this point is moot (in the sense of having little/no relevance). In the same vein as the former Treasury Sec’y testifying to Congress as ‘not being a regulator’, in contradistinction to his job design.

21stC American banking: its ‘Rafferty’s Rules’ – there aren’t any.

Posted by crocodilechuck | Report as abusive

I think someone’s misunderstood something here. The delay has gone into the drafting of the final rule, after consultation based on the Notice of Proposed Rulemaking. That doesn’t mean that the date of implementation of the rule itself is necessarily going to change – in the NPR this rule was meant to be slowly phased in and to take effect on 1 January 2018. Given that, I think it’s pretty unlikely that a slight delay to get the rule drafted correctly – or even a full Quantitative Impact Study – is going to change the implementation schedule.

Posted by dsquared | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/