I'm going to put forward two perspectives on why bad banks might be important in understanding the recession: an orthodox perspective; and a heterodox perspective.
Banks are financial intermediaries. A financial intermediary is a firm whose (primary) business is borrowing and lending. (They intermediate between the ultimate borrowers and the ultimate lenders). That's an important job. The ultimate lenders could lend directly to the ultimate borrowers. But then each ultimate lender would have to collect information on the creditworthiness or profitability of the ultimate lenders. Information is a non-rival good. There is no point in several ultimate lenders duplicating the costs of collecting that same information. A financial intermediary can collect the same information just once, and act as agent to the ultimate lenders' principal. To ensure the agent acts more or less in the principals' interest, banks act as residual claimant, and for this they need capital, so banks' capital can take the first hit when their loans go bad.
If banks go bad, because a lot of bad loans caused a loss of banks' capital, they must raise new capital, or shift to less risky loans, or contract their balance sheets in proportion to their loss of capital. Raising new capital generally means banks must issue new shares. The same financial crisis that caused banks' loan losses may also cause a fall in banks' share prices. People's uncertainty over the value of banks' assets may further lower banks' share prices. If banks believe that their share prices are below their fundamental value, they will be unwilling to issue new shares to raise new capital.
I'm not 100% sure my above story makes sense, either theoretically or empirically. But let's suppose it does. So we have "bad" banks, that have only their remaining capital, are unable or unwilling to issue shares to raise more capital, and so face a perfectly inelastic supply curve of capital -- they have what they already have, and cannot get more, at any "reasonable" price. They cannot expand their balance sheets to take advantage of what would otherwise be profitable opportunities, because each individual bank is capital-constrained. A fall in the interest rate on reserves at the central bank, which would normally give each individual bank the incentive to expand its balance sheet, even if it meant raising more capital by issuing new shares, will now have no effect. Because banks' balance sheets are capital-constrained. (See my previous post.)
So what. Why does this matter? That's what I'm trying to understand. If someone says "we can't get a recovery until bad banks get fixed", why is that?
If we view banks merely as financial intermediaries, then I find it hard to see why bad banks would really matter much, under present circumstances.
Sure, good financial intermediaries are important for the efficient allocation of the flows of savings between competing borrowers. They will make sure savings flow to the places where they earn the highest (risk-adjusted) return. A country with good financial intermediaries, doing this job well, will have a Long Run Aggregate Supply curve moving rightwards more quickly than an otherwise identical country with bad financial intermediaries, because the (risk-adjusted) return on savings and investment will be higher.
But the current economic crisis does not exhibit the symptoms of a problem with the AS curve. We have symptoms of excess supply of goods and labour, and falling prices. That suggests a fall in Aggregate demand, relative to Aggregate Supply. So even if bad banks have caused to LRAS curve to shift left (or move rightwards at a slower pace), that is not the current problem. We have a fall in AD, and current output and employment is demand-constrained, not supply constrained. We have a general glut.
Why should bad banks cause AD to fall?
Banks borrow short, safe, and liquid; and lend long, risky, and illiquid. And they earn their income on the spread between the interest rates. Bad banks will mean higher spreads, because capital-constrained banks are unwilling to expand their balance sheets to exploit and hence reduce those spreads.
Different people and firms face different costs of funds, or opportunity costs of funds. The whole 3D picture of the yield curve (across time, risk, and liquidity) will affect consumption and investment demand. "The" interest rate that determines consumption plus investment demand will be some sort of weighted average of the whole 3D curve. For a given interest rate set by the central bank, at the short, safe, liquid, end of the spectrum, the greater the spread(s) the higher will be "the" rate of interest which determines consumption plus investment demand. So bad banks increase the spread between the central bank rate and "the" interest rate, and so reduce Aggregate Demand, for any given rate set by the central bank. Bad banks could force the central bank rate down to zero, and still leave insufficient AD.
The above represents my best attempt to represent the "orthodox" view of why bad banks are (partially) responsible for the current recession. I think there is some truth in that orthodox view. I have proposed it myself in a previous blog post. And by "orthodoxy" here I mean the Neo-Wicksellian consensus that prevails in central banks, as well as academia, and views the monetary policy transmission mechanism as operating solely via interest rates. [And yes, Post-Keynesians, you are absolutely orthodox in that regard: horizontalist Neo-Wicksellians minus microfoundations. Sorry. I couldn't resist a little dig at your heterodox self-image ;-)].
But is that orthodox consensus the whole truth?
Financial intermediaries are useful (to reduce information costs) but not essential. A lot of lending and borrowing is direct, via stock and bond markets, and does not go through financial intermediaries. And banks aren't the only financial intermediaries either. I have often been a financial intermediary myself, when I both borrow and lend at the same time. (OK, sure, I miss on the technical definition of "financial intermediary" because my primary business is being an economics professor, but so what?).
The textbook definition of a bank is: a financial intermediary, some of whose liabilities are media of exchange. Banks create money, in that outdated viewpoint. The fact that people have chequing accounts at banks, and those chequing accounts are the most important means for people to buy and sell everything else, is what makes banks special, and important. Currency is the only alternative. And it's an outdated viewpoint because in the current orthodoxy the quantity of money is demand-determined. Monetary policy is interest rates, and interest rates cause aggregate demand, and if they also cause money, that is just as an aside.
Against that orthodoxy, the heterodox view argues that general gluts are always and everywhere a monetary phenomenon; the proximate cause of an excess supply of goods is always an excess demand for the medium of exchange. And the quantity of money (qua medium of exchange) is supply-determined, in a way that is quite distinct from any other good. It's a view arising from the confluence of disequilibrium monetarist and Keynesian streams.
Money is an asset (and usually a liability too), but is different from every other asset because everybody regularly both buys and sells money whenever they sell and buy anything else. That's what it means when we say that money is the medium of exchange. There is only one way to hold more land than you are holding now, and that is to buy more land. There are two ways to hold more money than you are holding now. One way is to buy more money (i.e. sell more other things); the second way is to sell less money (i.e. buy less other things).
An excess demand for land cannot cause a general glut of newly-produced goods, even though a desire to spend part of your income to buy land counts as desired "saving" according to standard definition. If everyone is trying to save in the form of land, so everyone wants to buy land, and nobody wants to sell, what happens? They fail, of course. Unable to spend their income on land, unless people decide to save their income in the form of money, they must decide to spend it on something else instead. Absent an excess demand for money, an excess demand for anything other than newly-produced goods must be frustrated, and people change their plans and decide to buy newly-produced goods instead. The inexorable logic of Say's Law dictates there cannot be a general glut that way, unless people desire to save in the form of money.
If people decide to save more, and save in the form of money, they might try to buy more money (i.e. sell more other things, like labour). They will fail of course, in a general glut, where all goods (like labour) are in excess supply. But an individual can never fail to to sell less money (i.e. buy less goods) and get more money that way. Each individual can succeed, but in aggregate they must fail (if the total stock of money is not increased). But the result is a fall in goods bought, and a fall in goods produced, and a fall in labour demanded.
A general glut is caused not by an excess desire to save, but by an excess desire to save in the form of the medium of exchange.
So where do banks come in? Banks create money. Sure, the central bank creates currency, but most people buy most things not with currency, but with the money created by commercial banks. We pay by cheque, or debit card, or ATM from our chequing account.
Forget currency (mainly, I admit, because I can't really get my head around where it fits in). Assume that all money is bank money. The commercial banks are the central bank's agents for creating money. Normally, when the central bank wants to create more money, it shifts the supply curve of reserves (shifts a horizontal supply curve down, or shifts a vertical supply curve right, or anything in between), and individual banks respond to this incentive of a lowered supply-price of reserves to expand their balance sheets, supplying more money on the liability side (and supplying more loans, buying more bonds, computers, whatever, on the asset side).
But even if banks want to supply more money, how can the actual stock of money increase, unless people want to hold more money? For any other asset, you cannot sell more unless people demand to hold more. If banks were land banks, they could not sell more land unless they persuaded people to want to own more land, by lowering the price of land, or raising rents, or whatever. But money is different. And it's different because it is the medium of exchange.
When a bank makes a loan, it does so by crediting the borrower's chequing account $100 in return for an IOU for $100. The borrower may ask about the interest rate on the loan. But he will not ask about the interest rate on his chequing account. Nearly all borrowers borrow money because they want to spend it, not because they want to keep it in their chequing accounts. Individual banks increase the interest rate on chequing accounts not because they want to persuade people to accept more loans, but only because they want to attract more deposits so they don't have to borrow as much from the central bank.
If the central bank increases the supply of reserves, and banks respond by increasing the supply of loans, the money supply expands, regardless of whether those borrowing from the bank demand a greater quantity of money. Banks really are like helicopters, in that they can increase the quantity of money held without needing to creating a demand to hold more money. For any other asset, if you wanted people to hold more, without getting them to want to hold more, you would have to give the asset away, for free. Throw it out of a helicopter.
Now commercial banks certainly don't give away money for free. But they can "force" people to hold more money even when people don't want to hold more money. Each individual borrower accepts money in exchange for his IOU, because he can get rid of that money by buying something. But in aggregate they can't get rid of the money they don't want to hold. One person's spending of money is another person's receipt of money. But their attempts to get rid of that money are what gets us out of the recession, by eliminating the general glut of other goods.
Central banks got rid of their fleet of helicopters when currency got replaced by demand deposits as the primary medium of exchange (again, I admit, I'm unclear on where currency fits into my picture). They contracted out helicopter operations to the commercial banks. They wanted the commercial banks to create the excess supply of money at current levels of income, leading people to want to spend that excess so we can escape any general glut. But now the banks' helicopters won't fly, because banks lack the capital to expand loans.
It ain't the loans what matter; it's the money creation that goes along with loan creation what matters. That's why (bad) banks matter. Bad banks won't create money.
Update: I wrote this post partly as a follow-up to my previous post on banks, and partly as a response to a post by Scott Sumner that was sort of a response to that post.