Wednesday, March 15, 2023

On marking to market and risk management

Two more thoughts:

1) In the SBV debacle, many of my colleagues and friends jump to the conclusion, we should just mark all assets to market and forget about this "hold to maturity" business.

Not so fast. Like all imperfect patches, there is some logic to it. Suppose you have a $100 payment that you have to make in 10 years. To cover that payment, you buy a $100 face value Treasury zero coupon bond. Done, zero risk. 

Now interest rates rise. The value of your asset has fallen in value! It's only worth, say, $90! Are you underwater? No, because when the time comes, you still will have exactly $100 to make the needed payment. 

You will quickly answer, well, mark both assets and liabilities to market. The $100 payment is now also worth $90, so marking both sides to market would reveal no change. But there is a lot of unneeded volatility here. And in most cases, the $100 payment is not tradeable on a market, while the $100 asset is. So now, you're going to be balancing marking to market vs. marking to model. Add the regulator's and many participant's distrust of market prices, which are always seemingly "illiquid," "distressed," in a state of "fire sale," "dysfunctional," and so forth. Add the pointlessness of it all. In this situation we all know that you can make the payment in 10 years. Lock it up and ignore it. Call the asset "hold to maturity." 

Of course, suppose the point of that asset is to make sure that depositors with $100 accounts can always get their money back by selling the asset. Well, now we have Silicon Valley Bank. 

Hence the imperfect fudge of current accounting and regulation rules. "Hold to maturity" assets don't get marked to market, and indeed there are penalties for selling them to meet current needs. Lots of "liquidity" and other rules are supposed to make sure there are adequate short run liabilities to stop a run. Those were of course completely absent in SBV's case -- a truly spectacular failure of elementary regulation. 

In short, mark to market makes sense to assess if a bank can make its payments and avoid failure tomorrow. Hold to maturity makes a bit of sense to assess if a bank can make its payments and avoid failure years from now, when both long term assets and long term liabilities come due. That is, if it survives that long. 

2) There is a lot of criticism of SBV bank management and board for being underinvested in risk management and over invested in lobbying, political connections, donations to politically popular causes, and so forth. Ex post, their choice of managerial investments looks brilliant! What brought in the millions to stem a run, I ask you? In today's highly political banking system, they made optimal choices. To an economist, many puzzling actions are just an optimal answer to a different question. 

Update: Ok, I went too far with that one. Management are out, shareholders wiped out. I'll stick with the idea that uninsured depositors did a great job of monitoring -- they monitored that the bank had the political chops to demand and get a bailout of uninsured depositors! 

From a correspondent: 

"It seems to me now that SVB was really a money market fund with the addition of a bit of equity and breaking all the SEC asset and liquidity rules that MMFs are subject to. " 

Or, it was really a mutual fund (money market funds with $1 values can't invest in long term bonds, long term bond funds must have floating NAV) that was violating rules on floating NAV! 



Small bank thoughts

 Three small thoughts. 

1) There is much commentary that bank troubles will interfere with the Fed's plan to lower inflation by raising rates. Actually, this is a feature not a bug. The main mechanism by which, in the Fed's view, raising interest rates slows the economy and lowers inflation is by "constricting credit," "tightening financial conditions," lowering borrowing that finances investment and consumer durables purchases.  The Fed didn't want runs, no, but it wants the result. If you don't like that, well, we need to think of other ways to contain inflation, like taking the fiscal gasoline off the fire. 

2) On uninsured deposits. A correspondent suggests that the Fed simply mandate that all large depositors participate in the sorts of services, there for the asking, that split large accounts into multiple $249k accounts spread over multiple banks, or sweeps into money market funds. 

I don't think that mandating this system is a good idea. If you're going to do that, of course, you might as well just insure all deposits and keep it simple. 

But the suggestion prompts doubt over the oft repeated notion that we want large sophisticated depositors to monitor banks. Anyone who was large and sophisticated enough to monitor banks had already gamed the system to make sure their accounts were insured, at some nontrivial cost in fees and trouble. The only people left with millions in checking accounts were, sort of by definition, financially unsophisticated or too busy running actual companies to bother with this sort of thing. Sort of like taxes. 

We might as well give in, that all deposits are here forth insured. If so, of course, then banks are totally gambling with the house's money. But we also have to give in that if they can't spot this elephant in the room, asset risk regulation is hopeless. The only workable answer (of course) is narrow deposit taking -- all runnable deposits invested in reserves and short term treasuries; fund portfolios of long term debt with long-term borrowing (CDs for example) and lots of equity.

3) Liquidity and fixed value are no longer necessarily tied together. I still don't quite get why better payment services are not attached to floating value funds. Then we wouldn't need run-prone bank accounts at all. 

 


Tuesday, March 14, 2023

How many banks are in danger?

With amazing speed and impeccable timing, Erica Jiang, Gregor Matvos, Tomasz Piskorski, and Amit Seru analyze how exposed the rest of the banking system is to an interest rate rise.

Recap: SVB failed, basically, because it funded a portfolio of long-term bonds and loans with run-prone uninsured deposits. Interest rates rose, the market value of the assets fell below the value of the deposits. When people wanted their money back, the bank would have to sell at low prices, and there would not be enough for everyone. Depositors ran to be the first to get their money out. In my previous post, I expressed astonishment that the immense bank regulatory apparatus did not notice this huge and elementary risk. It takes putting 2+2 together: lots of uninsured deposits, big interest rate risk exposure. But 2+2=4 is not advanced math. 

How widespread is this issue? And how widespread is the regulatory failure? One would think, as you put on the parachute before jumping out of a plane,  that the Fed would have checked that raising interest rates to combat inflation would not tank lots of banks. 

Banks are allowed to report the "hold to maturity" "book value" or face value of long term assets. If a bank bought a bond for $100 (book value) or if a bond promises $100 in 10 years (hold to maturity value), basically, the bank may say it's worth $100, even though the bank might only be able to sell the bond for $75 if they need to stop a run. So one way to put the issue is, how much lower are mark to market values than book values? 

The paper (abstract):  

The U.S. banking system’s market value of assets is $2 trillion lower than suggested by their book value of assets accounting for loan portfolios held to maturity. Marked-to-market bank assets have declined by an average of 10% across all the banks, with the bottom 5th percentile experiencing a decline of 20%. 

... 10 percent of banks have larger unrecognized losses than those at SVB. Nor was SVB the worst capitalized bank, with 10 percent of banks have lower capitalization than SVB. On the other hand, SVB had a disproportional share of uninsured funding: only 1 percent of banks had higher uninsured leverage. 

... Even if only half of uninsured depositors decide to withdraw, almost 190 banks are at a potential risk of impairment to insured depositors, with potentially $300 billion of insured deposits at risk. ... these calculations suggests that recent declines in bank asset values very significantly increased the fragility of the US banking system to uninsured depositor runs.

Saturday, March 11, 2023

Silicon Valley Bank Blinders

The Silicon Valley Bank failure strikes me as a colossal failure of bank regulation, and instructive on how rotten the whole edifice is. I write this post in an inquisitive spirit. I don't know the details of how SVB was regulated, and I hope some readers do and can chime in. 

As reported so far by media, the collapse was breathtakingly simple. SVB paid a bit higher interest rates than the measly 0.01% (yes) that Chase offers. It attracted large deposits from venture capital backed firms in the valley. Crucially, only the first $250,000 are insured, so most of those deposits are uninsured. The deposits are financially savvy customers who know they have to get in line first should anything go wrong. SVB put much of that money into long-maturity bonds, hoping to reap the difference between slightly higher long-term interest rates and what it pays on deposits.  But as we've known for hundreds of years, if interest rates rise, then the market value of those long-term bonds fall. Now if everyone comes asking for their money back, the assets are not worth enough to pay everyone back.  

In sum, you have "duration mismatch" plus run-prone uninsured depositors. We teach this in the first week of an MBA or undergraduate banking class. This isn't crypto or derivatives or special purpose vehicles or anything fancy. 

Where were the regulators? The Dodd Frank act added hundreds of thousands of pages of regulations, and an army of hundreds of regulators. The Fed enacts "stress tests" in case regular regulation fails. How can this massive architecture fail to spot basic duration mismatch and a massive run-prone deposit base? It's not hard to fix, either. Banks can quickly enter swap contracts to cheaply alter their exposure to interest rate risk without selling the whole asset portfolio. 

Saturday, March 4, 2023

Economic Journal Home Bias

Home Bias in Economics Journals is an interesting new paper by Dirk Bethmann, Felix Bransch, Michael Kvasnicka, and Abdolkarim Sadrieh (via Marginal Revolution).

...Researchers from Harvard, but also nearby Massachusetts Institute of Technology (MIT), and from Chicago (co-)author a disproportionate share of articles in their respective home journal.... We study this question in a difference-in-differences framework, using data on both current and past author affiliations and cumulative citation counts for articles published between 1995 and 2015 in the QJE, JPE, and American Economic Review (AER), which serves as a benchmark. We find that median article quality is lower in the QJE if authors have ties to Harvard and/or MIT than if authors are from other top-10 universities, but higher in the JPE if authors have ties to Chicago. We also find that home ties matter for the odds of journals to publish highly influential and low impact papers. Again, the JPE appears to benefit, if anything, from its home ties, while the QJE does not. 

On the bottom end as well, 

articles with a Chicago aliation in the JPE are less likely to be amongst the group of relatively low impact articles (i.e., to rank among the 25% or 10% of least cited articles published in the three journals in a year) than articles in the JPE authored by researchers from other top-10 institutions. 

Those are the what, but not the why. These findings naturally provoke some thought from my time at Chicago, and as JPE editor. 

Thursday, March 2, 2023

Fair tax full text

From the Wall Street Journal Feb 2. After 30 days I can post full text. 

A Consumption Tax Is the Shock Our Broken System Needs

Something remarkable happened last month. On Jan. 9, Georgia Rep. Buddy Carter introduced the “Fair Tax” bill to the House of Representatives, and secured a promise of a floor vote. The bill eliminates the personal and corporate income tax, estate and gift tax, payroll (Social Security and Medicare) tax and the Internal Revenue Service. It replaces them with a single national sales tax. Business investment is exempt, so it is effectively a consumption tax. Each household would get a check each month, so that purchases up to the poverty line are effectively not taxed.

Lessons from Sargent and Leeper

At the AEI fiscal theory event last Tuesday Tom Sargent and Eric Leeper made some key points about the current situation, with reference to lessons of history. 

Tom's comments updated his excellent paper with George Hall "Three World Wars" (at pnas,  summary essay in the Hoover Conference volume). Tom and George liken covid to a war: a large emergency requiring immense expenditure. We can quibble about "require" but not the expenditure. 


(2008 was a little war in this sense as well.) Since outlays are well ahead of receipts, these huge temporary expenditures are financed by issuing debt and printing money, as optimal tax theory says they should be. 

In all three cases, you see a ratcheting up of outlays after the war. That's happening now, and in 2008, just as in WWI and WWII. 

After WWI and WWII, there is a period of primary surpluses -- tax receipts greater than spending -- which helps to pay back the debt. This time is notable for the absence of that effect. 


We see that most clearly by plotting the primary deficits directly. The data update since Tom and George's original paper (dots) makes that clear. To a fiscal theorist, this is a worrisome difference. We are not following historical tradition of regular, full employment, peacetime surpluses. 


The two world wars were also financed by a considerable inflation. The important consequence of inflation is that it inflates away government debt. Essentially, we pay for part of the war by a default on debt, engineered via inflation.