Are the TBTF Banks Unnecessary?
October 23, 2011
I was having an interesting discussion with a friend who works at a Too-Big-To-Fail bank, and has for much of his career. He had some interesting insights which made sense to me, so I will try to pass them on.
He said, basically, that these institutions are inherently unnecessary. They don’t provide any useful service. In effect, they have become rentiers, sucking resources out of the economy, without providing anything meaningful in return. Indeed, they are worse than simply providing nothing, they are inherently destructive.
This probably sounds sort of Marxist, or something like that, but he meant this in a capitalistic sense. What is a bank for? What purpose does it serve? A bank is an intermediary for capital, accumulating capital from one group (“savers,” or basically bank depositors and buyers of bank debt), and distributing it to another, namely borrowers. These are your businesses and corporations, home mortgages, auto loans and so forth. On the investment banking side, the investment bank is supposed to raise capital for corporations by facilitating the sale of bonds or equity. In the secondary market, they facilitate the trading of existing bonds and equity.
That’s about it. Banks don’t do much more than that. And, these are pretty simple businesses. Banks have already been bypassed in large part on the corporate side, because even relatively small corporations are capable of issuing bonds directly today. Banks remain lenders to small businesses, and individuals.
Ideally, banks lend for new investment. Investment creates a new productive asset, anything from a pizza restaurant to medical training. The new investment creates the means to pay back the capital used, from new productivity. Borrowing for “consumption,” i.e. your credit cards and so forth, and arguably even auto loans, creates an obligation (debt to repay), but does not create an asset capable of repaying the loan. On a large scale, it destroys capital — capital is not directed into new productive assets, it is consumed. So, I would say that banks’ tendency to emphasize consumption-oriented borrowing in the last three decades or so is also a matter of overstepping their original purpose of channeling real savings into real investments. While investment is beneficial on a societal level, lending for consumption tends to be detrimental. Think about what things were like in the 1960s. Credit cards were unheard of. Borrowing money for consumption, like a vacation, was considered reckless. People went to college without borrowing money.
I say this to narrow even further what banks are supposed to do — their function in our capitalist society. They make loans ideally for productive investments, such as for business expansion. Maybe even home loans fall a little too close to the “consumption” side, although I generally think this is OK. Probably you are thinking: “Well, the availability of cheap mortages is essential for home ownership! Otherwise, we would be a society of tenants, and some rich guy would own everything.” That’s a reasonable line of thinking, but consider this: the homeownership rate in Mexico is 80%. Yes, 80% of the people in Mexico own their own home. Mexico, traditionally, hasn’t had much availability of mortage lending at all, due to currency instability, although this has changed somewhat in the last five years or so. Even at the peak of the housing bubble, in 2004, U.S. homeownership rate hit a historic high of 69.0%, and has been falling ever since. So, maybe mortgages reduce homeownership? Think about that.
Since mortgages are rare in Mexico, you can conclude that not many of those people who own their houses have a mortgage. Indeed, only about 13% of Mexican homes have a mortgage, compared to about 70% in the United States. Not only do Mexicans — average working Mexicans — own their homes, they really, really own them, free and clear. Probably their houses are smaller and simpler than ours. And that is a problem because … ? Is Mexico an outlier, some sort of one-off miracle? India has a homeownership rate of over 80%, and Tegucigalpa, Honduras has a rate of 79%. Owning your own home is the natural state of humanity, just like owning your own clothes. When you think about it, it makes sense. The Native Americans had a homeownership rate of 100%, and had no idea what a bank was.
So, maybe we don’t need banks for homeownership either. What’s left?
Since larger corporations are happy to issue bonds directly today, this leaves small businesses as a natural customer for banks. Anything else? Not much else, really.
Even business lending is a natural consequence of high corporate taxes. The most basic source of capital for business expansion is operating profits. Take those profits away via taxation, and there’s less left to reinvest, so you have to raise new capital from debt or equity. Second, because interest is paid before tax, debt is a much cheaper source of capital than equity. This produces an unnatural bias for corporations to rely on debt rather than equity financing. So, perhaps even small business borrowing is rather unnaturally emphasized today, compared perhaps to where it was in the 19th century when there was no corporate tax.
Banking, in the 1960s, was a pretty simple business. You made some loans to the people in your neighborhood. Indeed, since that time, banks’ traditional business has shrunk even further. I’ve already mentioned corporate disintermediation via the bond market. Also, the brokerage business has been commoditized, since the liberalization of trading commissions in the 1970s. Investment bankers still make their 6% on new deals. Banking today should be even simpler and smaller than it was in the 1960s.
What about securitization? One thing that has developed over the past thirty years or so is the idea of taking small bank loans, such as mortgages, auto loans, or small business loans, and packaging them into standardized securities so they can be sold like a bond issued by a single issuer. In principle, this might reduce banks’ role still further, since they wouldn’t have to hold onto the loan, they could just package it up and sell it on the public bond market. However, my friend reminded me that the events of the recent decade have shown the inherent problem with this structure. Banks use securitization as a way to basically scam the end buyer. This could be through excessive complexity, like CDOs or basically any product ending in an “O,” or simply by stuffing a bunch of bad loans in and saying they were good loans, which Bank of America and numerous other banks are being sued for today.
My friend says that, if anything, there is interest in moving away from securitized products and basically going back to plain old bank debt. Today’s investor who has been burned on his CDOs and CPDOs and RMBS and CMBS and every other acronymic product, is thinking: I’ll buy the debt of your bank, or a Certificate of Deposit which is basically the same thing, and you can keep the loan on your balance sheet. The rate might be a little lower, but you take the risk, instead of just trying to stuff me with a bunch of junk. The end-user of this sort of debt, like a pension fund or an insurance company, doesn’t have the expertise to make sense of a packaged loan deal. This is something that the banks have been taking advantage of. (These people want to act like they have the expertise, and banks give them lots of chances to let them act that way if they want to.) The problem, from a government perspective, is that your pensions and insurance companies tend to blow up.
So, once again we are trending, at least in principle, back to banking as it existed in the 1960s, or perhaps even an inherently smaller banking system because of the aforementioned disintermediation and liberalization of secondary trading commissions. As Paul Volcker put it, the only meaningful “financial innovation” of the last three decades was the Automatic Teller Machine. He wasn’t kidding. This means that we need even fewer bank employees than we had in the 1960s.
The traditional banking business, making loans, and the traditional investment banking business, primary issuance of securities and secondary trading — in other words, the services that actually provide some sort of benefit for society — are both simple, low-margin, shrinking businesses. Thousands of regional and neighborhood banks handle this business just fine. Hundreds of investment banks are capable of doing IPOs and other capital raising for companies. The local credit union will make you a home loan just as well as Bank of America. The big banks were driven to find new sources of revenue and profits. All of these new businesses were inherently parasitic. Proprietary trading, now evolved into systematic front-running. Stuffing people with consumer lending and credit cards. Stuffing institutional investors with “structured products” which were really scams. Derivatives — lots and lots of derivatives! Layer upon layer of asset management fees, none of which, in aggregate, can improve the average investment performance. None of this stuff existed in the 1960s, but the economy was healthier then than it has been at any time in the last forty years.
I have been promoting a reorganization of banks’ balance sheets that solves their problems with insolvency, but doesn’t particularly change their operations much. They would continue somewhat the same as before. But, after considering these arguments, I have to say it makes sense to liquidate these institutions completely, and re-regulate the industry so such things cannot happen again. Back to 1960s banking, but with ATMs and Ameritrade. We already have thousands of other banks and investment banks ready to take up any loan origination and securities-issuing needs.
March 23, 2008: How Banks Work 7: the Lender of Last Resort
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
For a while, I have been poking fun at the “fractional reserve banking” types, who I am quite sure have no idea what they are talking about. The real problem with “fractional reserve banking” is not some mysterious process of money creation or credit creation from nothing, but rather that levered banks tend to go bust rather regularly. My friend notes that U.S. banks have been going bust at a rate of about once a decade. I thought that was an exaggeration, but … actually, that’s about right. There were the busted sovereign loans in the mid-1980s, the busted real estate loans in the late 1980s, the busted lending to emerging markets in 1998, the TMT-lending bust in 2002 and of course the housing etc. bust today. Each time, the banks were bailed out by the IMF, “extend and pretend” (i.e. lying), stealing from the government, being allowed to do something that they should not be allowed to do (high-frequency trading), and so forth. Who needs it.
What could be done about this? Banks serve two functions: one is as a payments system, via their demand deposits. The other is as a way of aggregating a lot of small loans into a simplified product. Traditionally, this meant that the bank would hold the many small loans, and the investor would lend money to the bank in the form of CDs or debt.
The function as a payments system has already been partially replaced by money market funds. Money market funds are also problematic — there are liquidity issues — but, nevertheless, they have two nice attributes. The first is that they are financed entirely by equity. In practice, “breaking the buck” is so catastrophic for them that the consequences are akin to bankruptcy. But, nevertheless, they don’t really go bankrupt in a technical sense. Second, their liquidity is fairly well matched, with on-demand liquidity on one side and maturity of less than one year on the other. It would be pretty hard for this payments system to break down in any meaningful way. Even if a money market fund did “break the buck,” the potential losses are on the order of 5% or 10%, not a total wipeout. You could even require a very fat reserve, i.e. deposits at the Fed or similar, in the neighborhood of 20%, to resolve some liquidity issues. For many years, banks used reserve levels of this sort, so it wouldn’t be anything new.
The second function of banks, to aggregate small loans, can take place in somewhat the fashion that it already does, but separate from the payments system/demand deposit funding. Banks make small loans, keep them on the books, have adequate capital (10% seems to work), and finance themselves with debt of maturity of over one year. Although you could have various forms of securitization, there’s a lot to be said for having the loan originator and servicer being 100% liable for any losses. If the bank fails, the workout is fairly simple: you just sell off anything that you can get par value for, let the rest of the loan book run off, and pay back the creditors with the proceeds. This is separate from the deposits/payments system, so it should have no great systemic effects. These banks will be barred from any use of derivatives, which are important primarily as a way of hedging their duration mismatches. Since they would have a lot less duration mismatch, their need for hedging would be reduced. No off-balance sheet nonsense.
Perhaps there is a place for a new sort of structured product, which basically mirrors a bank: the bank/originator would take a 10% “equity” tranche, and the remaining 90% would be debt. No other tranches or quirks. Then, the bank would be on the hook for losses, but the effects of losses in excess of 10% to the structure would be relatively clear and simple.
These two functions could be combined operationally somewhat as they are now — retail branches could look about the same — and aggregated under a “bank holding company” umbrella. The idea is that the deposits/payments system would essentially be a money market fund, i.e. equity not debt, and be bankruptcy-remote from the other banking operations.
I’ve talked about a “Super Glass-Steagall” in the past, and we could develop those ideas a bit more here. Regular commercial banking is once again segregated from invesment or merchant banking. However, the broker/dealer format is also broken up. Trading in secondary markets and prime brokerage is separated from investment banking. All of the other activities of banks today, proprietary trading and so forth, are eliminated. You could still have derivatives, but they would have to be properly capitalized and in another separate, bankruptcy-remote entity. I imagine these derivatives operations would tend to go bust rather regularly, to the continuous disappointment of their counterparties and investors, until finally people realize that maybe they don’t need the hassle after all. No proprietary trading, even “dealing” or “market making.” If you want to be a dealer, then you can open your own fund.
But, even if you don’t like my various solutions, the fact of the matter is the U.S. banking system during the 1950s and 1960s, under the 1930s-era regulations, worked pretty well. You don’t have to be creative, just fix the problem.
The peasant uprising: My friend (same one) also had some interesting comments about the recent Occupy XXX movement. It is basically a peasant uprising. There have always been peasant uprisings. In a peasant uprising, the message is some variant of: “please don’t beat me so hard.” This is nice, and sometimes leads to gentler treatment of the peasants, for a while, but it generally doesn’t produce big changes. The same old masters adjust the beatings so they are just strong enough that the peasants keep working, instead of demonstrating.
He had an interesting insight. A successful change movement has a vision of what comes after the change. The old order passes, and a new one takes its place. The American Revolution began with a vision of independence from Britain. Many Marxist movements had a vision of a “worker’s paradise.” Gandhi in India had a vision of “India for the Indians.” The point is, we have a vision of what replaces the present status quo — even if, in the case of some Marxist movements for example, that vision is flawed.
One problem today is that the status quo in principle is actually quite good — government as described in the Constitution. One of the great advances in the science of oppression, compared to the cruder old techniques, is that the mechanisms of oppression are, in a sense, hidden and nameless.