The S&L crisis of the 1980s was centered around loans to real estate developers. The banking crisis of 2007-09 was centered around loans to real estate developers (not mortgage loans). Dodd-Frank was supposed to fix this problem. How’s it working so far?
This Financial Times story suggests the answer is “not well”:
Fears that smaller US banks are taking excessive risks in commercial real estate deepened on Friday as shares plunged in an Arkansas lender that has ploughed billions of dollars into developments in states as far away as New York.
Bank OZK — which this year changed its name from Bank of the Ozarks — revealed that it was taking a $45m writedown on two commercial real estate (CRE) loans, sending its shares down by more than a quarter in afternoon trading.
Bank OZK has $22bn in assets, half of which are CRE and multi-family real estate loans, making it a leading example of a nationwide trend: As big banks have pulled back from CRE and regulators have warned about high valuations, small banks have turned to the asset class as opportunities to expand in other areas have declined.
While many pundits on both the left and the right have focused on the issue of big banks and mortgage loans, I’ve consistently argued the real problem is smaller banks and commercial real estate loans. Consider these two questions:
1. What’s causing excessive bank lending on commercial real estate?
2. Why is it a public policy issue?
It turns out that the answer to both questions is identical: the Federal Deposit Insurance Corporation.
It’s literally indefensible to let banks lend taxpayer-insured funds to property developers. Full stop.
Large banks are more diversified than small banks, which helps to explain why they are less likely to take excessive risks. The fragmented nature of the US banking system also explains why Canada (with its large diversified banks) has not experienced the sort of financial crises that we frequently experience in America. If a few big property loans made by a big bank fail, the shareholders of the big bank usually absorb the loss. If a few major property loans made by a small bank end up in default, it’s very possible that the US taxpayers will bear a part of the losses.
If that Arkansas bank sounds vaguely familiar, it’s because I did a post discussing OZK Bank in June 2017. As long as we keep creating moral hazard with FDIC, these problems won’t go away. Dodd-Frank didn’t solve the problem for the exact reason that the “re-regulation” after the S&L crisis didn’t solve the problem. Policymakers refuse to do anything about moral hazard, because it’s so politically popular. Not just in banking, but also in federal flood insurance, health care, and many other areas.
People seem to want the government to create a giant featherbed for them, so they don’t have to worry about anything at all. But that merely shifts risks up to the macro level.
PS. Of course I’m not saying people should not have access to safe investments for their life savings. I have no objections to banks that invest FDIC-insured funds in safe investments.
READER COMMENTS
A
Oct 21 2018 at 4:38pm
Do you also support removing government interventions in the case of national financial crisis? It would be politically difficult to allow regional systems to fail, while national scales receive aggressive support, which historically flows through big bank balance sheets.
Scott Sumner
Oct 21 2018 at 5:48pm
A, I oppose all financial bailouts. Of course I also favor NGDPLT, including “whatever it takes” open market operations, so any financial crisis that did occur would not cause a deep recession.
Ahmed Fares
Oct 21 2018 at 6:54pm
Dr. Sumner,
“It’s literally indefensible to let banks lend taxpayer-insured funds to property developers. Full stop.”
The risks of property development are always incident on taxpayers. If not in the cost of bailing out failed banks, then in their capacity as consumers. If it becomes more difficult to fund property development, the higher costs will be passed along to consumers in higher prices for final goods and services.
Remember the BP oil spill? By requiring offshore drillers to bear the cost of carrying higher insurance, the cost of the marginal barrel of oil, and by extension all barrels of oil, rose. Consumers, not the oil company shareholders, paid for that higher cost of insurance at the pump.
In that latter case, the cost of cleaning oil spills is always incident on consumers. How consumers pay for that, whether by public funds for cleanups or higher pump prices is irrelevant.
Bank failures are like oil spills. We can only choose how the public pays for that.
robc
Oct 22 2018 at 9:33am
I have no problem, in my role as consumer, in paying higher prices for the insurance. As a taxpayer, it is immoral to have it foisted upon me.
Of course, the former is also a problem if the insurance is mandatory. A private FDIC-type insurance would be great. A regulation that requires the bank to have it, even if private, not so much.
Scott Sumner
Oct 22 2018 at 11:55am
Consumers should pay the full cost of any product, unless perhaps there are some positive externalities. That’s more efficient.
Matthew Waters
Oct 21 2018 at 9:49pm
Full-reserve banking is the ultimate answer. Who would determine whether FDIC funds are in safe investments? The ratings agencies? The simplest way forward is to end all public guarantees of bank assets (FDIC and discount window) with 100% reserve accounts.
You would have to pair full-reserve banking with truly robust NGDPLT. The monetary base would expand from roughly $5T to $15T should 100% of M2 be reserved.
I would want OMOs to go directly to consumers (say, offering to buy bonds up to zero maturity with 0% discount on coupons and principal). Then have helicopter money rather than OMOs on private assets.
Scott Sumner
Oct 22 2018 at 11:57am
Matthew, I see no need to go that far. Just have banks invest the insured funds in Treasury securities.
bill
Oct 22 2018 at 1:41pm
Could the banks invest in reserves held at the Fed? That pays higher interest rates than Treasuries.
Matthew Waters
Oct 22 2018 at 2:23pm
Yes, banks have to hold minimum reserves at the Fed and can hold excess reserves. Before 2008, excess reserves were near zero and required electronic reserves were only about $30B. The low required reserves ultimately supported trillions of dollars in payments. The banking system did extensive netting and short-term credit to minimize zero-yielding reserves.
Institutional balances also moved to money market funds and repos. The lack of interest on reserves brought about the “shadow banking system” IMO.
Since 2008, the Fed paid interest equal or exceeding T-bills. Today, electronic reserves held by banks at the Fed are about $2T. Through IOR, the Fed dramatically increased reserves and decreased the money multiplier. Unlike George Selgin, I think that’s fine.
But the Fed needs to open up access to those reserves. Selgin should look at the regulatory capture. Interest on reserves is a very good policy outside the zero-lower-bound, as it prevents a shadow banking system from forming. But then we should end the regulatory capture preventing charters (state, OCC, and Fed BHCs), and the Fed’s rejection of Master Account.
Matthew Waters
Oct 22 2018 at 2:10pm
I see 100% reserves as equivalent to 100% reserves and Treasuries. Since reserves earn interest, they’re functionally one-day T-bills. Treasuries still have interest rate risk and, well, why not go all the way? Things are simplified by $1000 in the bank truly being $1000. Meanwhile, I do not see efficiency benefits to $1000 being a mix of reserves and Treasuries.
The free market may already be providing this solution through The Narrow Bank, which is only bank accounts with 100% reserves. The New York Fed has slow-walked a Master Account application for over a year. According to TNB’s complaint, the denial came ultimately from Powell himself. But the New York Fed has not issued a final denial.
The New York Fed’s slow-walking of the pure regulatory capture by TBTF banks. The Fed can ultimately offset the increase in reserves from TNB’s deposits with buying more Treasuries. The fact that TNB threatens institutional deposits at TBTF banks can’t be separated from slow-walking the application.
Benjamin Cole
Oct 21 2018 at 11:47pm
Great post. Ever since 2008 I have pointed out that commercial real estate plummeted in value in a parallel drop to house prices, globally and in the United States.
For reasons I cannot explain, the public narrative and indeed even most macroeconomists have focused on housing markets.
Scott Sumner always had a wider perspective on the real estate price decline than others.
If bank bailouts are bad, we need to find another mechanism to keep the banking system solvent after real estate price declines. After all, even without FDIC insurance, banking systems can collapse.
robc
Oct 22 2018 at 9:34am
What wrong with the banking system collapsing?
That is just the market at work.
Matthew Waters
Oct 22 2018 at 11:58am
A bank’s solvency should not be a public policy concern. The “stability of the banking system” sounds good on its surface. Who is not against stability? But the arguments for public support and regulation fall apart under a robust NGDPLT regime.
The idea of fractional-reserve banking goes back to gold specie. Unlike fiat currency, gold specie was limited. When the true medium of exchange was specie, government support of fractional-reserve notes and deposits could control NGDP. Otherwise, there would be hard limits from mining.
With fiat currency, the arguments for fractional-reserve banking and maturity mismatch start going away. To meet NGDP goals, unlimited reserves can be printed.
Then the argument for publicly-backed banking becomes the bank deposit to investment conduit. In other words, full-reserve banking has higher C and lower I for same NGDP. I’m not sure if that’s correct. Even if it is true, does it matter? Much of the I from the banking system can be malinvestment.
S D
Oct 22 2018 at 5:37am
Excellent post Scott.
The price of real estate is also volatile, but even if the land value falls to zero, you still have a structure that people can live in. That’s why there is generally an upper limit on losses from mortgage portfolios as the property always has some value.
Lending to real estate developers is basically a pure bet on the price of land. The price of undeveloped land is always, everywhere volatile. The expected value over the lifetime of a development (from land acquisition to handover of keys) displays huge variance. It is much better suited to equity financing than debt financing.
It is something that the taxpayer should not subsidise and that supervisors should monitor very carefully.
LK Beland
Oct 22 2018 at 11:28am
Canada has a banking oligopoly. Thanks to reduced competition, banks are able to charge very high fees to manage savings and checking accounts, which offers them a steady stream of profit that insulates them from financial crises.
Another example: if I remember well, in 2013, the Canadian banks started to lower the rate that they charged for mortages, as the BoC was lowering it’s prime rate. The Minister of Finance officially asked the banks not to compete as hard and to stop lowering the rate charged to customers.
LK Beland
Oct 22 2018 at 11:43am
My memory served well. It was 2013:
https://www.theglobeandmail.com/real-estate/the-market/flahertys-decision-to-intervene-in-mortgage-war-raises-questions/article9949201/
As for Canadian banking fees, here’s an example:
https://www.cibc.com/content/dam/personal_banking/bank_accounts/pdfs/cibc-pasf-en.pdf
Fees are a lot higher than in the US. That explains a good deal of the system’s stability.
Alan Goldhammer
Oct 22 2018 at 2:36pm
@Scott or others who might know:
Is there a breakdown anywhere about the types and % of those that banks loan out? Clearly not all of their loans are to property developers (and even then housing developers are in a different league from commercial real estate developers). Banks make loans for cars, small and big businesses, home (and home equity), etc. It would be interesting to know what this breakdown is. I think Wells Fargo published a rough breakdown in their annual report but I sold the stock a while back and have not followed the filings.
Matthew Waters
Oct 22 2018 at 2:58pm
Bank assets are broken down here:
Fed H.8 Report
Of $12.8T in total bank assets, $2.2T is in Commercial Real Estate.
Scott Sumner
Oct 23 2018 at 1:16pm
Bill, I’d rather the Fed get rid of IOR. But yes, banks can invest in reserves.
Scott Sumner
Oct 23 2018 at 1:18pm
I recall that most bank failures during the Great Recession were caused by commercial loans.
Comments are closed.