“My objective today has been to review the evidence on the link between monetary policy in the early part of the past decade and the rapid rise in house prices that occurred at roughly the same time. The direct linkages, at least, are weak. […] Stronger regulation and supervision aimed at problems with underwriting practices and lenders’ risk management would have been a more effective and surgical approach to constraining the housing bubble than a general increase in interest rates.”
“Monetary Policy and the Housing Bubble”
Chairman, Board of Governors of the Federal Reserve System
At the Annual Meeting of the American Economic Association, Atlanta, Georgia, January 3, 2010
The Chairman of the Federal Reserve, Ben Bernanke, a former college professor and a stickler to economic data understands well how to navigate the treacherous waters of interpreting technical relationships between economic variables, especially when his talks are directed to his equally nit-picky colleagues at the American Economic Association (AEA). Econometrics, as a science, has no methodology to conclusively establish causality between variables. Directional causality can, at best, only be inferred within a margin of error, provided the econometric equation is constructed well by selecting the appropriate variables, which is a choice the modeler makes.
Still, the question remains: Does money supply cause (housing or other) bubbles, because if not money what can, independent of where that money came from ― the central bank or from foreign savings ― into U.S capital markets? To answer this question, it would be useful to shed light on the uses of the money being supplied by the central bank. Bernanke’s response to this in his speech to the AEA was that the central bank should have regulated better to direct the money toward better or better managed uses. This answer, even though it accurately establishes the complementary nature of the relationship between base money supply and following that money once it has been supplied, however, is not complete.
If the same question he was attempting to answer in his paper is rephrased as: “Does money supply cause financial asset prices to rise?” the perspective of analysis changes because the housing assets were financial assets in the value chain. And financial assets are a form of broad money. The question then becomes: “Does base money supply, which the Fed and/or the capital markets provide, raise the broad money supply?” The housing crisis was a result of the appreciation of the broad money supply or the derivative financial assets which were created with cheap base money, whichever route it took after first originating within the United States.
Money was initially lent at very low variable rates to both kinds of homeowners: those who could afford and to those who could not afford their homes once the rates adjusted. The ratio of all the broad money in the economy to the real annual output or GDP is an economy’s financial depth. U.S financial depth or leverage before the crisis was 10 dollars for every real dollar or $150 trillion for an economy which was really worth $15 trillion. The Fed is paying back that money now with one real dollar for every 10 leverage or notional “dollars” (or the debt ― principal and interest upon maturation ― accounted for as the total current and expected assets on the balance sheets of all the financial institutions in the value chain, each claiming its share of the total matured debt), because the leverage was all financial: money was invested to buy “money.” This recapitalization was necessitated because the base of the housing pyramid began shaking after some homeowners could not pay on their borrowings and the banks could not recover the cash they had given out.
If for a second, we assume that housing debt was not securitized and no financial assets, besides the loans on the books of the commercial banks, were created out of household investment ― a component of the country’s gross domestic product ― would money supply have raised stock values? Probably, as is evident from the NASDAQ bubble which burst in 2000. This implies that cheap money can raise the prices of financial assets and housing enabled the creation of those financial assets.
The volume of housing-related financial assets would only go up if homeownership is increased. The prices of homes will rise if the demand for homes rises. Demand will rise if more money is lent out toward homeownership, whatever the source of that money, domestic or foreign. Meaning, the most likely motivator for the financial markets to increase homeownership was the attractiveness of real homes to produce a value chain of financial assets, just as more public corporations increased the total volume of stocks being traded in the financial markets which had created the boom in initial public offerings (IPO) in the late ‘90s to benefit from the stock price boom after having created more of them first.
I would have constructed the econometric equation more comprehensively to examine the relationship between base money and the various measures of broad money and some proxy variable for the effectiveness of regulation (realizing that data quality is a constraint and itself a function of the regulatory structure), to study both monetary relationships and the relationship between money supply and regulation. Then, it is obvious ex ante that whatever the strength of the relationships between the variables as revealed by the estimation, there would always be a firm causal relationship between one form of money and another. If base money causes the prices of financial assets to rise or expands the broad money supply, which it does, then it caused the housing bubble at least to some degree because all homes are ultimately financial assets.
The Chairman of the Fed could not be comprehensive and instead had to rely solely on the Taylor Rule (a very indirect macroeconomic relationship that the Fed uses as a baseline for monetary policy which does not include variables for either bank health or the regulatory structure) because the Fed does not measure broad money. Bernanke would have been more accurate had he attempted to answer if it was feasible to establish the Fed’s intent to create a housing bubble by keeping interest rates low in the period 2000-2003, because the explicit purpose of expanding money supply in that period was to provide liquidity to an economy that was subject to the popping of the tech bubble, corporate governance crisis and 9/11/2001. Econometrically, it is not possible to establish the normative intent of policy to create bubbles, besides the known objective function of the Federal Reserve to achieve its mandate.
The question then becomes: why did the Fed raise rates during the housing bubble solely based on the deviation of inflation expectations (and its inflation forecast) from its implicit inflation target which it does not formally announce like the Bank of England (BOE) and the European Central Bank (ECB)? What was the Fed’s inflation target to know the deviation about which the Fed was so concerned to implement an interest rate policy of staying on the steady path of raising rates to normalcy in the period 2003-2007? Further, if the Fed is concerned about inflation expectations beyond its tolerance over its forecast horizon, did the Federal Open Market Committee (FOMC) consider the context of those expectations which was the housing bubble, particularly in the period 2005-2007? If it did, did the FOMC carefully evaluate the larger weight it placed on inflation expectations over the possible fallout from the collapse of the bubble. Controlling bubbles through interest rate increases had so concerned Alan Greenspan that he avoided raising rates as a matter of policy during suspected bubbles. If the FOMC did try to evaluate the economic consequences of a bubble collapse, was the analysis and data from the then FOMC Vice Chairman Timothy Geithner’s New York Fed and other market sources about the effects of interest rate increases on the secondary markets independently validated, given the data constraints associated with that data because those markets are not regulated by the Fed?
The crisis was the result of excessive leverage and the ensuing poor quality of broad money due to an inadequate regulatory structure. There is a direct causal relationship between money supply and financial market bubbles, the housing bubble being one of them. The issue for verification is the strength of that direct relationship, ceteris paribus, to understand better how the regulatory structure must be changed going forward so as not to leave out any financial market activity to raise the quality of broad money that the markets create.
Sound money, whether it is the cash the Fed supplies or its derivatives created by the financial markets, is the best safeguard against financial and economic crises.