Scott Sumner  

American banking: Socialism or laissez-faire?

PRINT
U.S. Immigration Has Had a Lar... Cook vs. Cass on Global Warmin...

Imagine a banking system where the public would lend money to the Treasury at X% interest, and the Treasury would then lend the same funds to commercial banks at the same X% interest rate. All bank deposits were held by the Treasury. That sounds kind of socialistic, doesn't it? And yet this might also be viewed as a description of the actual American banking system, during the 1990s and 2000s. The Treasury backstops all FDIC-insured deposits.

Of course in recent years, progressives have complained that the 1990-2007 system was a product of radical deregulation, and the 2008 banking crisis was blowback from public policies that reflected a laissez-faire ideology. So which is it?

I'd say neither. Ideological framing gets in the way of understanding what's actually going on in banking. After a recent post criticizing FDIC, commenters suggested that removing deposit insurance would hurt small savers. But if protecting small savers were the goal, surely there are much less costly ways of doing so.

While waiting out endless delays at the airport, I recently came across a magazine that specialized in commenting on major New Your City real estate developers. (No, this story has no Trump angle.) This article caught my eye:

Sometimes it seems like Bank of the Ozarks is the only show left in town, lending hundreds of millions to developers across the city amid a condo financing drought. Its aggressive approach has led some industry insiders to question whether the Arkansas-based lender is becoming overexposed to a downturn in the New York City condo market. . . .

Ozarks, which was for most of its history a small community bank with a handful of branches in Arkansas, has been one of the most prolific lenders in the city in recent years. It recently provided a $108 million construction loan for Xinyuan Real Estate's new condominium project at 615 10th Avenue in Hell's Kitchen. It's also a lender on JDS Development and the Chetrit Group's Brooklyn megatower rental project at 9 Dekalb Avenue and on Tishman Speyer's Macy's development in Downtown Brooklyn. . . .

Last year, Carson Block, founder of Muddy Waters Research, said he was shorting Ozarks' stock because he believed the bank was moving too aggressively in the real estate space.


I don't believe in "bubbles", so don't take this as a prediction that Bank of the Ozarks will run into trouble. Rather, my point is that this sort of aggressive strategy makes far more sense in a world of FDIC, than without deposit insurance. How many people would want to put their life savings into this sort of bank, if FDIC did not guarantee the deposits?

Back in the 1920s, banks were managed far more conservatively than today. Patrick Sullivan recently reminded me of a study by Eugene White, which pointed out that the huge 1920s real estate boom and bust did not cause a banking panic:

Although long obscured by the Great Depression, the nationwide "bubble" that appeared in the early 1920s and burst in 1926 was similar in magnitude to the recent real estate boom and bust. Fundamentals, including a post-war construction catch-up, low interest rates and a "Greenspan put," helped to ignite the boom in the twenties, but alternative monetary policies would have only dampened not eliminated it. Both booms were accompanied by securitization, a reduction in lending standards, and weaker supervision. 'Yet, the bust in the twenties, which drove up foreclosures, did not induce a collapse of the banking system. The elements absent in the 1920s were federal deposit insurance, the "Too Big To Fail" doctrine, and federal policies to increase mortgages to higher risk homeowners. This comparison suggests that these factors combined to induce increased risk-taking that was crucial to the eruption of the recent and worst financial crisis since the Great Depression.
Many people are surprised to find out that banks were managed much more conservatively during the 1920s than today. After all, weren't there frequent bank failures prior to the enactment of FDIC in 1934? Yes, but this reflected two special factors:

1. Because of unit banking laws, the US had thousands of small, poorly diversified banks in rural areas. Many of these banks failed during the 1920s and 1930s.

2. Monetary policy was far more unstable during this period, resulting in a roughly 50% fall in NGDP between 1929 and early 1933, as well as other big declines in 1920-21 and 1937-38. In contrast, NGDP fell by only 3% during the Great Recession.

Screen Shot 2017-06-01 at 10.26.17 PM.png
The importance of the unit banking laws is obvious when you consider than Canada had no bank failures during the Great Depression, despite a similar fall in NGDP. Today, that sort of decline in NGDP would wipe out virtually the entire US banking system. Canada owed its success to having large well-diversified banks with branches all across the country.

So why doesn't Congress fix the moral hazard problem? Because for Congress, moral hazard is not a bug, it's a feature. Cloud Yip recently interviewed Charles Calomiris, who had this to say about moral hazard in banking:

The pattern is that the governments, on the one hand, protects banks in the forms of deposit insurance or government bailouts when a crisis happens, or give banks certain opportunities. In exchange, the government asks things from the banks. In the last 40 years, especially across democracies, the thing that the government was asking the bank to do is to get heavily involved in residential real estates funding or to subsidize mortgages.

One of the thing that has been happening all over the world is that people in democracies are hoping that their governments can come up with programs that would make housing more affordable. The easiest way for governments to do that is to get banks to subsidize mortgage risk. The way the governments get the banks to subsidize mortgage risk is by protecting the banks. They give the banks something in exchange and then tell the banks "OK. Now we have given you the protections. We have given you these new rights. They are very valuable to you. The cost is that you have to do something that we, the government, found politically expedite." What Sophia Chen and I are finding now is that the story about the US in our book "Fragile by Design" may actually be a broadly-based story.


There are many possible solutions to moral hazard. For instance, my checking account is a Fidelity MMMF, invested in safe government bonds. We don't need FDIC to offer safe investments to small savers. Another option is requiring all FDIC-backed deposits to be invested in save assets such as Treasuries, and let banks use uninsured deposits for riskier loans. In that case we could basically repeal all of Dodd-Frank. Indeed, we wouldn't need any bank regulation at all. But that sort of reform would hurt bank profits, especially at smaller banks. And they have far more political power than America's taxpayers, who will again be on the hook when another round of go-go bankers runs into trouble.

PS. You don't see either party trying to fix the moral hazard problem. There's a reason for that.


Comments and Sharing






COMMENTS (11 to date)
Paul T writes:

Interesting article, thanks.

Can you perhaps link to more discussion on your position "I don't believe in bubbles", if you have blogged about this before?

Antischiff writes:

Dr. Sumner,

Yes, and it's telling that the big banks have recently said they want to keep most of Dodd-Frank in place.

Colin W writes:

I always felt the main benefit of deposit insurance was to prevent runs on banks with solvent but illiquid balance sheets, but customers didn't know the state of the bank (which is why they withdrew their deposits). If we got rid of deposit insurance I think the key question we'd have to address is whether or not depositors would have enough information to know the solvency of their bank/could effectively monitor their bank.

bill writes:

Something to consider on this: "Another option is requiring all FDIC-backed deposits to be invested in save assets such as Treasuries, and let banks use uninsured deposits for riskier loans."
I wasn't able to find a good source for the total of FDIC-backed deposits, but I think I found that FDIC-backed banks hold $11 trillion in deposits. Can that be right? If it is, there just aren't enough T-bills to back the deposits. And T-notes and T-bonds have too much duration risk. Besides, I'd rather see the Treasury reduce our debt over time.

Andrew_FL writes:

Apart from equivocating on calling the US banking system socialized, this is the best piece you've done in quite some time.

I have a question, though: do you know if the data on Canadian GDP is available online back to the Depression?

Scott Sumner writes:

Paul, I have many such posts; here's a very recent one.

http://www.themoneyillusion.com/?p=32492

Antischiff, Yes, the paperwork burden is much worse for their smaller competitors.

Colin, Canada didn't have it until 1966, while Australia didn't have any until 2008. AFAIK, their banking systems did fine.

Bill, I certainly don't want to insure all $11 trillion in deposits. Rather we should protect small depositors. Wealthier people like me have no trouble finding safe places to put their money.

Andrew, Thanks. I don't know about Canadian data. The FRED site has a section on historical data, you could look there.

Ralph Musgrave writes:

The last para of the above aricle says: "Another option is requiring all FDIC-backed deposits to be invested in safe assets such as Treasuries, and let banks use uninsured deposits for riskier loans." Basically that's full reserve banking, which I support.

At least four Nobel laureate economists also supported that, including Milton Friedman, James Tobin, Merton Miller and Maurice Allais. Re Friedman, see for example his book "A Program for Monetary Stability", especially 2nd half of Ch3.

Scott Sumner writes:

Ralph, As I recall, under full reserve banking all insured deposits are invested in cash, not bonds.

Jake writes:

Nice article Prof Sumner.

In my view the banking system is socialist for the banks.

It will never change of course since under fractional reserve the banks get to turn a $1 deposit into $10 in loans. No wonder they are so rich and powerful.

I support full reserve since it dispenses with the fiction that a dollar can be two places at once -- loaned to someone else, but also on demand in your checking account. Get rid of that and bank runs are no longer an issue, eliminating the FDIC.

T-bills are probably an acceptable substitute for cash though.

Follow up question -- wouldn't a full reserve system make monetary policy somewhat easier for the Fed to control?

Scott Sumner writes:

Jake, You asked:

"Follow up question -- wouldn't a full reserve system make monetary policy somewhat easier for the Fed to control?"

That was Friedman's view, but I think it applies more to the gold standard than to fiat money. I suppose it might help indirectly by reducing the risk of financial crisis, as monetary policy is more difficult to control during a financial crisis.

Ralph Musgrave writes:

Scott, Re your querie (June 3, 2017 10:13 AM) there are different versions of full reserve. Most advocates go for the cash option. But Milton Friedman favored the bond option (i.e. government bonds). But both cash and government bonds are extremely safe, so there isn't much difference between the two options.

Comments for this entry have been closed
Return to top