When I was just out of graduate school, I received an offer to work in the mortgage backed securities department of a major Wall Street firm. When I visited, they were quite happy as they had just made a killing in the mortgage market. They had noticed that households tended to refinance mortgages early in some months more than others, subtly affecting the value of the mortgages they were buying and selling. I remember being impressed by the power of leverage. When you borrow $36 for every dollar you have, a one percent gain (36 cents) becomes a 36% return on your dollar. They turned a small mispricing into a large gain. But now we are seeing the downside of leverage. Losses are also magnified.
To understand the crises, one first needs to understand how home buyers get mortgage loans (and how students get student loans). Suppose Microsoft finds that, after paying bills today and collecting payments from customers, it has a few million dollars left over at the end of the day. Now suppose that, rather than put the cash in a checking account which earns no interest, Microsoft loans the money overnight to Lehman Brothers on what is known as the repo market. Lehman in turn buys your mortgage (or student loan) from a bank. Microsoft has then financed a mortgage. More than half of all mortgages are now financed this way, rather than being financed by bank deposits, as used to be more common. Tomorrow, when Microsoft wants it’s money back plus interest, so Lehman again borrows in the Repo market to pay off Microsoft, essentially rolling over the repo loan.
This complex arrangement has some advantages. First, traditional banks hold mostly mortgages from a particular region and not other investments. If the economy suffers in that region, the bank may face huge loses. Lehman, which invests in many regions and many assets should be able to offset any mortgage loses with, say, gains in oil futures. To further diversify risk, Lehman may take out insurance which pays off if the mortgage is not repaid from an insurance company like AIG.
Now, the crises. Suppose repo lenders become concerned that Lehman will not be able to repay the overnight loan, because the mortgage is in default and not making payments to Lehman. Suppose the repo market refuses to lend to Lehman. Since Lehman must pay Microsoft the next day, Lehman can go bankrupt quickly. It has borrowed to make an investment that has gone sour, and thus owes more money that it has in terms of mortgage assets. AIG may also go bankrupt like any other insurance company if insurance payouts exceed the value of it’s assets.
A general shut down of the repo market is what everyone wants to avoid. Picture only half of UM’s students being able to get a student loan, for example!
So many solutions have been proposed or already tried that I cannot address them all, so let me focus on the current bailout. Suppose Morgan Stanley sells it’s bad mortgages to the government, then uses the cash from the government as collateral to again receive loans in the repo market. Repo lenders would be happy to lend to Morgan Stanley, which now has no bad mortgages, and thus no possibility of bankruptcy. Morgan may be more careful about their leverage and loan quality, given what happened to Bear Stearns and Lehman (on the other hand Morgan might continue making risky loans with high leverage, knowing another bailout is possible). Repo lenders who received mortgages when Lehman went bankrupt would also benefit by selling the mortgages to the government. The only loser is the taxpayer. The FED and Treasury are hoping for this best-case outcome.
But before we commit $700 billion (!) to the bailout, it is worth asking what would happen if we did nothing. At the end of the day Microsoft must do something with it’s money. Perhaps Microsoft would simply put the money in conventional bank deposits. If so, conventional banks would be flush with cash and could increase lending. However, interest rates would be higher, since it was the investment banks and their innovative products which diversified risk and lowered the cost of borrowing for millions. But conventional banks are less leveraged, and so the resulting system may be more stable, without any cost to taxpayers. Indeed, two investment banks have already converted to conventional banks, so this may be the result even with a bailout. So it seems that repo lenders, investment banks, and marginal borrowers (who all took on too much risk anyway) lose if we do nothing, and conventional banks benefit.
Other solutions may be more beneficial in the long run. Previous rules that only three companies can rate mortgages have been lifted, which should result in more diversity of opinion. The U.S. has the second highest tax rate on saving in the world. Reducing taxes on savings would result in more household saving, and thus more cash being available for lending. Regardless of the outcome, I am glad I am studying the problem as an economist, not as an employee in the mortgage-backed security industry.