How Banks Work 7: The Lender of Last Resort

How Banks Work 7: The Lender of Last Resort

March 23, 2008


Last week, we talked about what happens when a bank, or a similarly levered financial entity like a brokerage, SIV, certain types of hedge funds, etc., has a “bank run.” In other words, lenders (depositors) want their money back. The bank has to either borrow from someone else, or sell assets.

March 16, 2008: How Banks Work 6: Liquidy Crises and Bank Runs
March 9, 2008: How Banks Work 5: Selling Loans
February 24, 2008: How Banks Work 4: Banks and the Economy
February 17, 2008: How Banks Work 3: More Elephant Poop
February 10, 2008: How Banks Work 2: Shitting Like an Elephant
February 3, 2008: How Banks Work

If nobody wants to loan them money, and they can’t sell assets effectively, then they go to the Lender of Last Resort, the central bank.

This is actually close to central banks’ original purpose. They were never intended to manipulate interest rates and manage currencies. That is something that happened later, after the Great Depression, when economists convinced themselves that manipulating interest rates and fooling with currencies would be helpful.

Central banking was originally developed, in the mid-19th century, to alleviate the liquidity-shortage crisis. This was accomplished by making loans, effectively increasing the supply of base money. The characteristic symptom of a systemic liquidity-shortage crisis is high interbank interest rates. Very high. Typically over 10%, when 3% might be a normal figure, and often above 40% during the crisis. Today, central banks’ interest rate targeting mechanisms effectively keep something like that from occurring, so the classic 19th century-style liquidity crisis doesn’t really exist any more. You can read about it in Chapter 8 of my book. 1907 was the last liquidity-shortage crisis in the U.S. Interbank rates went over 100% — for best quality credits.

The risk of having to pay such a high penalty rate, or perhaps not having access at all to funds, is the reason that banks held much larger reserves in those days, on the order of 10% of deposits. Banks today can accomplish a similar thing by holding very liquid securities, such as Treasury bonds or perhaps deposits/loans with other banks.

Central banks haven’t really used direct lending for many decades. They discovered that they could keep interest rates from spiking via open market operations. This was done in conjuction with the gold standard, but it was on the slippery slope to interest rate manipulation and floating currencies. The advantage of the old system of direct lending was that there was typically a rather harsh penalty interest rate, of 10% perhaps. Banks would only borrow at that rate in dire circumstances. And, when things settled down, they quickly paid the money back. This is the “self-cancelling commercial bill of credit” you sometimes read about in old economic discussions. The involvement of the central bank in the monetary system was rare, brief and self-reversing.

One advantage of the system of open market operations is that it leaves questions of solvency to the market. The central bank can buy a Treasury bond, paying for it by “printing money.” The money from this sale typically ends up at a bank — although, you should notice that the money was not directly given to a bank. It was given to the seller of the bond, in trade for the bond. This additional base money is often lent out on the interbank market, thus creating a tendency toward lower interest rates. However, who is it lent to? That is up to the receiving bank to decide. To someone solvent, presumably, rather that to an insolvent entity.

An entity can remain in business even if it is insolvent, if someone keeps lending them money. You can probably identify many households with that characteristic. The bankrupt S&Ls, like those of Texas, stayed in business for years in the 1980s because FDIC deposit insurance kept people from withdrawing their deposits.

The central bank was never meant to prop up weak institutions. Sometimes it is difficult for a lay person to recognize the difference between bank illiquidity and insolvency, but bankers understand well. Typically, an institution will be weak for months or years, due to deteriorating asset quality (bad loans etc.), before they suffer the final collapse in liquidity (“bank run”).

Thus, in the Great Depression, banks which were weak — those that had made too many real estate loans during the Florida property bubble for example — were expected to go under. After all, they didn’t know they were in the Great Depression. That label was applied many years later.

After the Great Depression, it was determined that the government should have some role in preventing a “systemic collapse.” Probably they should have “prevented” it by not hiking taxes to the moon. Central banks were not supposed to be government agencies, but rather something like a banking industry association, or simply a dominant commercial bank like the Bank of England was. Since central banks were already in existence to “make loans to banks,” to serve as a “lender of last resort”, etc. etc., this new government role was laid upon the existing central bank organization.

I am not particularly opposed to government support of the financial system to prevent “systemic collapse.” (I put it in quotes because it is complicated to define or describe, but everyone knows what I’m talking about.) Typically, as in Japan or the U.K. recently, or in many other countries with banking problems, this has taken the form of a direct government loan to the bank. In the U.S., however, people seem to look to the Federal Reserve for this role. By making a loan to the bank, the government acts as the Lender B we described last week. Note that this is a loan. It is supposed to be paid back, just like a commercial loan. If it is not paid back, the bank is bankrupt just as if it defaulted on a commercial borrowing. Thus, ideally, “taxpayers” get all their money back. The government’s role is to step in when fears of insolvency prevent a significant financial institution from finding another lender. The government turns something of a blind eye to asset deterioration, in recognition of the risk of “systemic collapse.” In the midst of high drama, it is hard to tell what assets are worth anyway. The Fed, via its discount window lending and other such direct facilities (we seem to get a new one every week these days), is also able to turn a blind eye to asset deterioration. This isn’t really “nationalization” either, as the government doesn’t have any ownership in the bank (equity). The bank would be nationalized, at least partially, if the government recapitalized it. However, even this is not really a “bailout” in the sense of free money. Existing shareholders would be diluted, and likely take a loss overall.

This lending doesn’t really do much for insolvency, alas. Lending affects the Liabilities side of the balance sheet, but not the Assets side. If assets are deteriorating, lending can’t help. However, it prevents a fire sale of assets, and also takes care of short-term liquidity issues, which alleviates the “systemic” risk. If the bank has troubles but manages to right itself, it eventually pays back the goverment loan and continues as a regular bank. If the bank is unable to right itself, then it has an “orderly liquidation” or perhaps a sale and merger with another entity.

The Fed has invented many different types of “loans,” including repo agreements and various swaps. They generally function much like loans, however.

Unlike the government, or other commercial entities, the Fed actually creates fresh money when it makes a loan. In many cases, this fresh money is likely “sterilized” via the interest-rate targeting mechanism, so that overall base money doesn’t necessarily change much. In effect, the Fed’s assets show an increase in lending, and a decrease in securities related to open-market operations, mostly government bonds. Indeed, the recent swap facility accomplishes this rather directly. Roughly 50% of the Fed’s capacity to make direct loans/swaps/repos/etc. has apparently been used thus far, for a total of $400 billion of such activities on an $800 billion-ish balance sheet. The Fed can’t exceed its $800 billion-ish without resorting to direct money-printing. It would be rather exciting if we got that far.

All in all, I don’t think the Fed has done anything particularly unseemly thus far regarding its lending activities. Probably a rather good job, given the situation. The problem is that, while doing so, the Fed is allowing the currency to become a big problem. Also, the Federal government should at least ready itself to take a bigger role, perhaps by recapitalizing financial institutions, or by making additional loans if the Fed’s capacity to do so is exhausted.

Notice that we haven’t once mentioned either interest rate manipulation or changes in currency value. The Lender of Last Resort function is independent of these later functions, which didn’t really come to full bloom until after 1971. Thus, we see that the role of the government/central bank (it should really be the government instead of the central bank) in preventing “systemic collapse” does not require a floating currency or interest rate manipulation, or even a central bank. Nor are these functions contrary to a gold standard. A gold standard just keeps the value of the currency stable. That’s all it does.

Later, we will talk about floating currencies and interest rate manipulations, and how those are also outgrowths of the Great Depression.

* * *

Bailout Watch: If there’s anything happening that most resembles a “bailout,” it is the defacto nationalization of Fannie Mae and Freddie Mac last week. Fannie and Freddie plan on buying $200B up to “trillions” of mortgage paper. Doug Noland of has the details. Like other closet Mogambo fans, he seems to share a fondness for Dramatic Capitalization!

* * *

People talk about the S&L “bailout.” S&Ls weren’t “bailed out.” They were liquidated. However, the depositors were FDIC guaranteed, up to a certain point. Thus, the depositors got their money back. It was the depositors that were bailed out. There was a lot of thievery and fraud going on at the time too, mostly within the government (and the Bushies were heavily involved), to pocket a piece of this bailout for themselves. Then, the S&L assets were liquidated. This was another great chance for certain insiders to make a bucket of easy money. They just sold themselves loans/assets that were worth perhaps $0.50 or $0.20 for $0.02. I believe they even got easy government financing to buy these loans. Thus, they had $0.02, but bought $5.00 of loans for $0.20, using 10x leverage. This is pretty much like hitting the lottery, except that no luck is involved. Bankruptcy and liquidation are great times to make a lot of money, without much risk, when nobody is noticing.

* * *

Interesting little essay on similar themes here:

The Shadow of the Depression and the Lesson the ’70s

* * *

Nice series of articles on naked shorting of smallcap stocks, by the CEO of, Patrick Byrne. This is some serious research by a smart guy, who has experienced this first hand.

* * *

Meanwhile, back at the ranch … MBIA and Ambac are still without plan or rescue (although Ambac has a little more capital), and the housing/mortgage market continues to deteriorate. Now, we’re starting to get economic issues like job loss and declining revenues. I think that this year, the housing market is going to start seeing some “give up.” Prices are still way too high. We could get a “dislocation” to the downside.

The Subprime Crisis is Just Starting