The Fed Terminates QE, We Lower Our GDP Forecast
by Paul Kasriel and Asha Bangalore
If you have been following our commentaries in recent months, you know that we have been putting a lot of emphasis on monetary financial institution (MFI) credit as a cyclical determinant of domestic demand for goods and services. To refresh your memory, we define MFI credit as the sum of the credit extended by the Federal Reserve, the commercial banking system, the savings and loan system and the credit union system. MFI credit is akin to what the Austrian school of economics would call created credit. That is, MFI credit is credit figuratively created “out of thin air.” The Austrians (and we) make a distinction between created credit and transfer credit. In the latter credit concept, spending power is transferred from the grantor of this credit to the recipient of the credit. Transfer credit, then, is funded by the grantor by postponing some spending that the grantor otherwise would have engaged in. Thus, an increase in transfer credit generally does not result in a net increase in total spending in the economy. Rather, it results in a change in the distribution of spending. Because created credit is funded “out of thin air,” an increase in it generally does result in a net increase in total spending in the economy.
We estimated the growth in real GDP using a linear least-squares regression with explanatory variables of real MFI credit growth, lagged real GDP growth and a constant. The actual and estimated values of the year-over-year percent change in real GDP are shown in Chart 1. Not too shabby for private sector work in our humble opinion.
Chart 2 shows the year-over-year percent change in real MFI credit from Q1:1953 through Q4:2010, the latest complete data available. In Q4:2010, the year-over-year change in real MFI credit was 1.67%, well below the median change of 4.58% over the entire period.
Although complete MFI credit data are available only on a quarterly basis from the Fed’s flow-of-funds report, which is released with a considerable lag, weekly data are available for the Fed’s and the commercial banking system’s contributions to MFI credit. Might we be able to get a more timely impression of the behavior of MFI credit by examining the weekly behavior of the sum of Federal Reserve and commercial banking system credit? The data shown in Chart 3 indicate we can. Chart 3 shows that since the late 1980s, the sum of Federal Reserve and commercial banking system credit as a percent of total MFI credit has been steadily rising. For example, in Q4:1988, the sum of Fed and commercial bank credit accounted for approximately 64% of total MFI credit. By Q4:2010, Fed and commercial bank credit had risen to approximately 87% of total MFI credit. So, the sum of Fed and commercial bank credit can be viewed as a reasonable proxy for total MFI credit now.
The Fed resumed quantitative easing at the beginning of November 2010. As shown in Chart 4, in the 22 weeks ended March 30, 2011, the sum of Federal Reserve and commercial bank credit had increased at an annualized rate of 5.2%. In contrast, in the 22 weeks ended October 27, 2010, just before the commencement of QE2, the sum of Federal Reserve and commercial bank credit had increased at an annualized rate of just 0.1%. So, coincident to the resumption of Fed quantitative easing, growth in the principal component of MFI credit accelerated.
What has been the primary contributor to the acceleration in this principal component in MFI credit growth since the commencement of QE2 – Fed credit or commercial bank credit? Fed credit. Chart 5 shows that in the 22 weeks ended March 30, 2011, Fed credit increased by a net $329 billion whilst commercial bank credit contracted a net $81 billion. Unless the commercial banking system steps up its credit creation in the second half of 2011, the mid-year termination of Fed quantitative easing implies a significant deceleration in the growth of nominal MFI credit. In turn, a significant deceleration in the growth of MFI credit implies a significant deceleration in the growth of nominal GDP. This is the principal reason we are lowering our second-half 2011 real GDP forecast.
A secondary reason for lowering our second-half real GDP forecast is the higher inflation currently present in the economy. The higher inflation, all else the same, lowers the growth rate of real MFI credit. Thus, even if growth in nominal MFI credit were maintained in the second half of 2011 at the same rate as the QE-enhanced first-half rate, the higher inflation rate would reduce the second-half real MFI growth rate, which would have negative implications for second-half real GDP growth. We should add with regard to the outlook for inflation, the relatively low current rate of growth in nominal MFI credit and the even slower anticipated growth in MFI credit with the termination of QE suggests that inflation is likely to moderate from its current pace rather than accelerate, barring, of course, further near-term negative supply shocks – droughts and/or energy production disruptions, for example.
Lastly, another headwind for the U.S. economy in the second half of 2011 is the potential adverse effect on U.S. exports emanating from monetary policy tightening actions in key emerging market economies. For example in China, the largest emerging-market economy, nominal M2 money supply growth has slowed sharply in recent months as one of the central bank’s policy interest rates has been ratcheted higher (see Chart 6). This would suggest that the pace of Chinese economic activity is likely to moderate soon, which would adversely affect the rest of the world’s exports, including those of the U.S.
Consistent with our reduced second-half 2011 growth forecast for real GDP, we also are lowering our interest rate forecast. We had been expecting the Fed to begin lifting its policy interest rates in January 2012. We now have pushed that out to April 2012. Because the yield on the Treasury 2-year security is very sensitive to expectations of near-term Fed policy interest rate decisions, we have lowered our 2011 forecast for this interest rate. Because the yield on the Treasury 10-year security is very sensitive to the rate of growth in real GDP and because we have reduced our real GDP forecast, we also have lowered our 2011 forecast for this interest rate.
We want to emphasize that this month’s forecast should be viewed as a transition forecast. That is, we do not yet know if the commercial banking system will step up significantly its credit creation in the second half of 2011 to counteract the cessation of Federal Reserve credit creation. If, in the next couple of months, we see no evidence of a pick-up in commercial bank credit creation, we would be prepared to lower further our real GDP growth forecast and push out beyond April 2012 the Fed’s commencement of policy interest rate hikes. In sum, we believe the risks are rising that the pace of second-half U.S. economic activity will disappoint risk investors, primarily because of the cessation of Fed quantitative easing at mid-year.
*Paul Kasriel is the recipient of the Lawrence R. Klein Award for Blue Chip Forecasting Accuracy
THE NORTHERN TRUST COMPANY
ECONOMIC RESEARCH DEPARTMENT
SELECTED BUSINESS INDICATORS