You can’t control what you don’t measure.
In engineering, control theory is all about using information gained by measuring a system to plan and carry out intelligent actions that will control it. Ideally, it leads to desirable outcomes, such as a nuclear reactor that doesn’t melt down, or a robotic arm that does precisely what it is supposed to do.
In the case of the economy, we might not be measuring everything we need to achieve control.
For at least half a century, policy makers seeking to control inflation and unemployment have typically focused on managing interest rates. The U.S. Federal Reserve lowers its target rate if the economy stalls, and raises it if inflation appears on the horizon. The recent financial crisis, however, had more to do with the amount of borrowing people did and the way such leverage fueled a bubble in the housing market.
Where does leverage currently fit in the equation of macroeconomic stability? Surprisingly, the answer seems to be that it doesn’t.
For 15 years, Yale economist John Geanakoplos has argued that policy makers should pay more attention to leverage. Prevailing interest rates determine the cost of borrowing, if a borrower ultimately repays the loan.
Leverage -- reflected by how much collateral people or firms need to put down to borrow and might lose if they fail to pay the loan back -- determines how much someone can buy with a given amount of starting capital. It’s an independent quantity that also influences what happens in the economy, whether borrowing is easy and attractive or not.
Importantly, leverage isn’t a fixed quantity. It changes over time as people get more or less optimistic and lenders more or less confident of being repaid. Geanakoplos makes a convincing case that such changes in leverage can and routinely do drive major economic booms and busts, and that managing leverage should be as much a part of the Fed’s activities as managing interest rates.
The core of his argument rests on a common-sense insight: An increase in leverage generally leads directly to an increase in prices. Consider the housing market. At any moment, you’ll find that some people are more optimistic than others, more convinced that prices will go up in the future and hence eager to invest on that belief. If banks change their practice to require only a 5 percent down payment, rather than 20 percent, these optimists have more to spend. They can purchase up to 20 times the value of their own funds, rather than just 5 times, hoping to profit when things go well. As more money chases the available houses, prices go up, which makes the optimists even more bullish.
In a series of papers and presentations, Geanakoplos has documented how the leverage effect operates in the real world (see further discussion on my blog). From 2000 to 2005, for example, down payments on home mortgages fell from about 15 percent to 3 percent. Leverage increased similarly for banks and hedge funds borrowing to buy mortgage-backed securities. Housing prices rocketed up, spurring home-building and all kinds of other economic activity.
The boom also set the stage for collapse. As markets grew more volatile, and everyone more uncertain, lenders of all kinds naturally wanted to preserve their money, so they increased collateral demands accordingly. By 2008, investors could buy only $1.20 in mortgage securities for each dollar of their own money, compared with $15 in 2006. The down payments required on new mortgage loans rose to as much as 30 percent.
Geanakoplos argues that this dynamic is not a one-off peculiarity of the latest financial crisis, of modern banking, deregulation and derivatives. He sees it as a natural cycle -- the leverage cycle -- that is fully able on its own to drive an economy up and down even if interest rates stay the same. He suggests we’ve had three crises linked to the leverage cycle in the past 20 years: in 1994, in 1998 and in 2008.
If you take his view seriously, it looks as if economic theory, and the Fed policy based upon it, hasn’t been paying attention to the right variables. Some economists are working to include leverage in the basic models that central banks use, and we can hope they will succeed. The Fed is beginning to collect data of the kind that would be useful for identifying leverage changes and their effect on economic activity, but how far leverage control will enter practical policy and regulation on banking and investment firms remains unclear.
Crises incur real damage, in the form of lost jobs, credit-starved companies and people who can’t borrow to buy a car or fix their house. Banks wouldn’t be in such precarious condition, and so many homeowners wouldn’t be underwater, if leverage had not reached such extreme proportions before falling back again. To avoid such outcomes in the future, we need to find ways to manage leverage as well as interest rates.
Since the crisis, there have been myriad proposals for new measures to detect systemic risk. The mathematics and data used are getting ever more complex. Yet one of the most important measures may be one of the simplest. We’ve understood the benefits and potential dangers of leverage for a long time. Now we need to start measuring it.
(Mark Buchanan, a theoretical physicist and the author of “The Social Atom: Why the Rich Get Richer, Cheaters Get Caught and Your Neighbor Usually Looks Like You,” is a Bloomberg View columnist. The opinions expressed are his own.)
To contact the writer of this article: Mark Buchanan at firstname.lastname@example.org.
To contact the editor responsible for this article: Mark Whitehouse at email@example.com.