by Alex J. Pollock
Looking back to 1960, there were in the United States: 13,126 commercial banks and 18,962 depository institutions. By the end of 2016, depositories totaled only 5,913, of which 5,113 were commercial banks. That’s a 69 percent reduction in the number of depository institutions, and a 61 percent reduction in commercial banks.
On the other hand, in 1960, the total assets of the commercial banking system were only $256 billion. Though hard to believe, the entire banking system had total assets of only about one-tenth of today’s JPMorgan-Chase, and only 1.6 percent of today’s banking assets of $15.6 trillion. Citibank — which wasn’t Citibank then, but the dignified First National City Bank of New York — had less than $9 billion in assets. To our minds, now muddled by decades of constant inflation — including a central bank that has formally committed itself to creating perpetual inflation — these all seem like very small numbers.
Instead of measuring in nominal dollars, to see through the fog of long years of inflation, we can measure banking assets consistently relative to the size of the economy, as a percent of annual gross domestic product. The $256 billion of commercial banking assets in 1960 was 47 percent of the $541 billion in GDP.
The increase is striking: by 2016, banking assets had gone from 47 percent to 83 percent of GDP. That is more than a 75 percent increase in the banking system’s size relative to the economy, at the same time the number of banks fell by more than 60 percent. At present, this ratio is close to its all-time bubble peak.
What is driving this growth? It’s not commercial and industrial loans. On the trend, their percent of GDP is flat at 8 percent to 10 percent since 1960. On average, the commercial and industrial loans of the banking system have kept up with the growth of the economy, but not more.
The real driving factor is real estate credit. The commercial banking system’s real estate loans rose relentlessly from 5 percent of GDP in 1960, to more than 26 percent at their bubble peak, and are now at 22.5 percent.
Nor is this the whole real estate story. With the popularization of mortgage securitization, the banking system’s securities portfolio, not only its loan portfolio, shifted to real estate risk. Going back to 1992, the sum of banks’ real estate loans and mortgage-backed securities as a percent of GDP has risen to 32 percent — six times the 1960 level.
In short, the vast bulk of the dramatic increase in the size of the banking system relative to the economy comes from the acceleration of real estate exposure — a rising trend for more than six decades. How can the banks keep doing this? Well, it helps to have your liabilities guaranteed by the government, both explicitly through deposit insurance and implicitly through bailouts and central banking.
Should the banking system keep getting bigger relative to the economy, and should this increase continue overwhelmingly to reflect real estate risk? That is a dubious proposition. As Columbia University’s Charles Calomiris has written (in a not-yet-published paper): “The unprecedented pandemic of financial system collapses over the last four decades around the world is largely a story of real estate booms and busts. Real estate is central to systemic risk.”
Very true. But as Calomiris notes, the Financial Stability Oversight Board, set up as part of the Dodd-Frank Act to oversee the U.S. financial system, “seems to be uninterested.”
The Bank Holding Company Act of 1956 defined a bank as an institution that accepts demand deposits and makes commercial loans. Neither part of this old definition still touches on the main point. A bank now is for the most part an institution that makes real estate loans and funds them with government-guaranteed liabilities.
This banking evolution poses a huge systemic question: How do you deal with a banking system whose risks are concentrated in real estate prices and leverage? To this question we are, as yet, without an answer. Do the supposed systemic thinkers at the Financial Stability Oversight Council even understand the magnitude of the historic shift in risk? Maybe a future FSOC with new members will do better.