Arnold Kling  

What I'm Saying

PRINT
The Historical Roots of the Fi... Why Get Insurance?...

If all goes according to schedule (unlikely), then about the time that this is posted, I should be in the middle of saying something along these lines:

I was a kid when the Beatles came to America. Soon they were followed by the Rolling Stones, the Dave Clark Five, the Kinks, and other groups. The press dubbed it the British Invasion, and it was a big deal for about a year and a half. It was everywhere in the media, with saturation coverage. Yet in spite of all that coverage, at that time I had no idea that the biggest influence on that music, the real root of that music, was African-American rhythm and blues. Because although the media coverage was a mile wide, it was an inch deep. There was no effort to understand the history or the evolution that produced the music.

I feel the same way about the financial crisis. There has been a lot of coverage, but most of it acts as though history began in 2008. There is no attempt to trace the history of how and why the financial system evolved to the point where there was a crisis. My paper is an attempt to remedy that.

In my talk today, I will just pull out a few points from the paper. Mostly, I want to talk about the implications going forward. Once you understand the history, I think you'll realize that today's proposals for financial reform are just a continuation of the self-defeating policy approaches that got is in trouble. We need to think differently if we want to create effective reforms.

This crisis may seem like it came from nowhere. A year and a half ago, who had heard of credit default swaps, collateralized debt obligations, or structured investment vehicles? I would say that even well-informed people, including most journalists and most people on the Hill, had never heard of these financial instruments before the crisis. So it's natural to assume that the regulators did not know about them either, that these new financial instruments and new financial relationships were created while nobody was looking.

But regulators were aware of these new financial instruments and new financial relationships. Regulators were aware of these innovations. They approved these innovations. They collaborated with banks in shaping how these instruments were used. And regulators put pressure on banks and created incentives for banks to use these instruments.

Let's look at the anatomy of the crisis. The crisis consisted of four elements: bad bets, excessive leverage, domino effects, and 21st-century bank runs.

Houses were bought that should not have been bought. Prices were paid for houses that should not have been paid. Mortgage loans were made that should not have been made. Financial institutions took positions in mortgage credit risk based on faulty assumptions. Those were bad bets.

The major financial institutions--the commercial banks, Freddie Mac and Fannie Mae, investment banks like Lehman and Merrill Lynch--all took too much risk with too little capital. That was excessive leverage.

When one firm got in trouble, there was a sense that if it failed the result would topple other firms. Those were domino effects.

Finally, these institutions were holding long-term mortgage assets funded with short-term instruments, such as repurchase agreements and commercial paper. When lenders began to doubt the value of the assets, they refused to roll over loans. This produced the modern equivalent of a bank run.

If you look at the history of how the financial structure came to be so fragile, you will see that it was the result of policy, particularly in two key areas--housing policy and bank capital regulations.

In housing policy, on Capitol Hill for some reason it became standard to equate the concept of affordable housing with mortgage credit that was subsidized and lenient. That was a self-defeating policy. Putting people in debt up to their eyeballs ultimately does not promote home ownership. What happened was that subsidized mortgage credit and lenient mortgage credit promoted speculation. In 1995, five percent of mortgage loans were backed by non-owner-occupied houses. By 2005, fifteen percent of mortgage loans were backed by non-owner-occupied houses. Speculation had tripled. Speculation drove up home prices, which made houses less affordable, which made the Hill call for more mortgage subsidies and more leniency, which fueled more speculation, and so on. It seemed like an endless cycle, except that it did end--with a crash.

Bank capital regulations were even more self-defeating. There was a continual increase in regulatory capital arbitrage, which means that banks found ways to take the same risk while holding less and less capital while staying within the rules. It was a continual increase that I describe in the paper, but let me highlight two dates. First, we have 1989, when the Basel accords were first implemented in this country. This was a set of risk-based capital standards that was supposed to prevent a recurrence of the S&L crisis. It was an attempt to ensure safety and soundness. They created risk buckets which gave a big break to Fannie Mae and Freddie Mac securities. That is, the capital requirements for those securities were low, which made it more profitable for banks to hold those securities than to hold mortgage loans that they originated themselves.

The next date is January 1, 2002, when the break given to Freddie Mac and Fannie Mae securities was extended to any security that could earn a AA or AAA rating. That is, the regulators outsourced bank capital regulation to the credit rating agencies. This set off a frenzy of innovations designed to manufacture AA and AAA rated securities out of junk mortgage loans.

The net result was that bank capital regulations, which were supposed to promote safety and soundness, did the opposite. There was excessive leverage, because the banks were able to hold the same risk with less and less capital needed to comply with regulations. The methods used to create this regulatory capital arbitrage involved tight interconnections among firms, creating domino effects. And the funding mechanisms for these new financial structures involved short-term lending, making the system vulnerable to 21-century bank runs. Thus, all of the main elements of the financial crisis were policy-driven, because of self-defeating housing policy and bank capital policy.

So where do we go from here? I think that if you understand the history, you can see that a lot of the proposals that are on the table for reform are just a continuation of the same self-defeating policies that got us here. We need new thinking. Here are three ideas.

First, stop equating affordable housing with mortgage subsidies and mortgage leniency. Instead, let the market determine down payment requirements, interest rates, and other mortgage terms. Use other ways to assist people who need help with managing their finances and with affording decent housing.

Second, start thinking about getting the mortgage-backed securities market disconnected from the feeding tube of government support. As you will see if you read the history, mortgage-backed securities were invented by government and have always required special government privileges. We need to think about disconnecting the government feeding tube. What will happen if we disconnect the mortgage securities market from the government feeding tube? My guess is that the market will die. And if it dies, then mortgage interest rates will be a little higher--studies suggest about one quarter of a point. Somewhat less capital will flow into mortgages, and there will be more capital somewhere else. But whatever benefit there is from having more capital in mortgages rather than somewhere else, that benefit has to be tiny compared to the risks involved and the catastrophe that has just occurred.

Third, we have got to continue to look for ways to make failure a viable and credible option for financial institutions. We're seeing a little of that, with phrases like "new resolution authority" and "living wills." Unfortunately, most of what we are seeing fits in with the self-defeating approach of trying to make banks too regulated to fail. But too-anything-to-fail is a self-defeating policy. When people assume that the regulators are going to prevent failure, then they have less incentive to prevent failure themselves. Russ will talk more about this in his presentation.

What I would like to see is a paradigm shift, where we stop talking about preventing failures with a systemic risk regulator and the like. Instead, we talk about business continuity planning. A year ago, we heard that if we didn't bail out companies like AIG, then next morning we would find that the ATM's wouldn't work. I do not understand how that can be. When I go to the ATM, I don't go there to buy a credit default swap.

If in fact the failure of one financial institution could cause ordinary transactions to be interrupted, that needs to be fixed. We need to create procedures and rescue teams so that if a bank goes under, the next day we can come in make sure that the ATM's still work, the checks still get cleared, and the credit card still works when I swipe it at the grocery store. If we have backup systems and business continuity plans in place, then we should be able to continue ordinary financial transactions regardless of whether a particular bank stays in business or not.

If we had a business continuity strategy, then allowing a bank to fail would be a credible and viable option for regulators. That in turn would change everyone's incentive to make them watch out for their own risks and watch out for their own exposures to counterparties. Instead, the more you try to make the system too regulated to fail, the more careless everyone becomes in their risk exposures. It is a self-defeating approach.

So, to wrap up, my recommendations are:

--Above all, read the history and learn the larger lessons.

--Stop equating affordable housing with mortgage subsidies and mortgage leniency

--Start thinking about disconnecting the mortgage-backed securities market from the feeding tube of government support

--Continue to look for ways to make failure a viable and credible option at big banks.


Comments and Sharing






COMMENTS (6 to date)
Josh Gessler writes:

Your analysis and history is very useful. I haven't read the paper yet (and maybe you address it there from a historical perspective), but from your summary, there is one critical factor I think your missing: embedded moral hazard.

You have these two paragraphs:

"The major financial institutions--the commercial banks, Freddie Mac and Fannie Mae, investment banks like Lehman and Merrill Lynch--all took too much risk with too little capital. That was excessive leverage.

When one firm got in trouble, there was a sense that if it failed the result would topple other firms. Those were domino effects."

Right after these paragraphs, there should be inserted:

"And, all of the players had a sense that if one of the big firms failed, or there was a systemic shock, that governments would intervene to protect the firms (or, at least, their bondholders and stockholders) and make counterparties whole. Otherwise, none of the major financial institutions would have done business with one another, given the excessive leverage of the firms and their vulnerability to cascading collapse."

Yancey Ward writes:

I find it hard to believe that removing the government "feeding tube" would raise mortgage interest rates only a quarter point.

Philo writes:

"We need to create procedures and rescue teams so that if a bank goes under, the next day we can come in make sure that the ATM's still work, the checks still get cleared, and the credit card still works when I swipe it at the grocery store. If we have backup systems and business continuity plans in place, then we should be able to continue ordinary financial transactions regardless of whether a particular bank stays in business or not."

I take it "we" is *the government*. Its present "business continuity plan" is, ultimately, socialization: if systems fail, the government will (as a last resort) take over and run things. This at least has the virtue of simplicity. You make it sound like the government should anticipate *every possible kind of failure* and construct a specific plan for each kind. But that would be too complex and costly.

David Beckworth writes:

Arnold:

I know you discuss the Fed's low interest rates in the early-to-mid 2000s as a contributor to this crisis, but it seems to get little coverage there. I think it plays a more important role--though not the only one--than you let on.

First, by lowering interest rates the Fed created problems for fixed income managers: they were not getting they returns they needed. There was this popular phrase "the search for yield" back then that described their attempts to find yields in places they normally may not have gone (eg. CDOs).

Second, by lowering rates for to historically low levels and for a prolonged period the Fed encouraged substitution out of conventional mortgages into exotic ones with low introductory teaser rates. (See Lawrence H. White on this point.) Thus, it wasn't just financial innovation but a distorted price signal (i.e. inordinately low rates) created by the Fed that encouraged home buyers to try new types of mortgages.

Third, given that many emerging markets peg their currency to the dollar the Fed's monetary policy got exported across the globe at this time. Even the ECB and the BoJ had to be mindful less their currencies lost export competitiveness and thus they loosely followed the Fed. The Fed's policies then helped create a global liquidity glut that fueled a global housing boom. Some of those funds got recycled back to the United States See here for more.

tjames writes:

Philo - such business continutiy plans do not have to be particularly complicated by a multitude of possible failure modes, since there really aren't that many failures the government need concern itself with. The 3 main failures that concern the goverment would be liquidity-only failures, solvency-only failures, and failures of liquidity and solvency at the same time.

Foremost in dealing with any of these is the need to determine which crisis you have, since your response may differ depending. Therefore, the first part of any plan must be the methods and resources to quickly determine this.

The liquidity-only failure has been addressed in a number of earlier blog posts - in a liquidity crunch, "government should lend freely, but at a penalty rate". The business continuity plan in this case is to be able to extend the credit quickly.

For solvency-only failure, I think Arnold is endorsing more clearly delineated plan for ordering of creditors and resolving the matter and doing so quickly, so as to limit solvency problems elsewhere in the system. Also of note is that in a situation - like mortage backed securities - where there are a lot of long term assets, short term liabilities, and very high leverage ratios, insolvency can very quickly become illiquidity as your creditors become aware of the situation. That leads to..

Liquidity and solvency failures simultaneously. I think the current crisis had both of these as you had instituions becoming more clearly insolvent, then becoming rapidly illiquid. The problem here is sorting and treating the 2 problem seperately. You want the ATMs to keep running and check clearing to continue(liquidity), but you do not want to have to bail out all the bad bets just for that purpose. Goverment had no plan to sort these these functions and support one without supporting the other, so did the only thing it thought it could, which is throw so much money at the situation that the ATMs stayed on. But, this means a lot of creditors and shareholders, who rightly should bear the burden of failed enterprise, are also getting at least partially bailed out.

One other aspect of this plan needs to be a contingency for dealing with this in a systemic way. In other words, the plan must address the domino affect Arnold talks about.

I won't pretend the details are going to be easy, but this is what Bernanke, Geithner, et. al. signed up for when they took the posts they did.

Greg Ransom writes:

Arnold, I don't understand how Fed interest rate policy didn't effect this:

"these institutions were holding long-term mortgage assets funded with short-term instruments"

Maybe you've explained it somewhere. But I've missed it.

Comments for this entry have been closed
Return to top