Mortgage Rates Remain Low

Information provided by the Federal Housing Financing Agency (FHFA) indicates that mortgage rates on purchases of newly built homes fell by 5 basis points over June to 3.93 percent. At this level, rates remain below the 4.18 cycle peak level recorded in February. Meanwhile, a more commonly used rate reported by Freddie Mac indicates that mortgage rates rose in October. Despite some monthly divergence, the two series track each other.

A previous post demonstrated that the 30-Year fixed rate mortgage can be decomposed into the rate on the 10-Year Treasury security and the residual, the mortgage risk premium. At the same time, the rate on the 10-Year Treasury security can be decomposed into the real return and a measure of inflation compensation. As shown in the figure below, which uses the Freddie Mac mortgage rate, the increase in the Freddie Mac mortgage rate over October largely reflected the rise in the real return portion of the 10-Year Treasury security, which rose by 13 basis points to 0.50 percentage points over the month. Meanwhile, inflation compensation increased 4 basis points to 1.86 percentage and the mortgage risk premium fell by 7.9 basis points to 1.54 percentage points. However, the mortgage risk premium typically remains steady over time.

The significant increase in the real return portion of the 10-Year Treasury security and its impact on mortgage rates was not a surprise. The Federal Reserve began its balance sheet normalization process in October, and analysis of the “taper tantrum” found that the increase in the 10-Year Treasury security in response to the taper tantrum reflected a rise in the real return portion of the 10-Year Treasury security. Mortgage rates were expected to rise modestly, partly reflecting an increase in the real return portion of the 10-Year Treasury note rate.

The real return portion of the 10-Year Treasury security is important because it may provide information about the growth rate of the economy’s potential level. While the time series is short, following the evolution of the real return portion of the rate on the 10-Year Treasury security (i.e. the actual rate on the 10-Year TIPS) over time, suggests that it tracks the estimate of actual r* by economists at the Federal Reserve Bank of San Francisco. The close pattern reflects the notion that the real return on the 10-Year Treasury securities is a measure of expected r*. Trends related to r* are important because r* is a key input into a rules-based approach to determining the federal funds rate.

In theory, r* is also related to the potential level of the economy because, as described by researchers at the Federal Reserve Bank of San Francisco, it “is the inflation-adjusted, short-term interest rate that is consistent with full use of economic resources…”. Since expected r*, the real return associated with 10-Year Treasury securities, tracks the estimate of actual r*, which is partially predicated on the economy’s potential, then expected r* should also track the growth rate of the economy’s potential level. According to the Congressional Budget Office, the growth rate of the economy’s potential level is expected to reach 1.8 percent on a 4-quarter basis by 2019.

Going forward, since the Fed target of 2 percent inflation only leaves room for an additional 14 basis points increase in the inflation compensation portion of the 10-Year Treasury note rate (as of the end of October), then significant increases in the nominal rate on the 10-Year Treasury note must be fueled by growth in expected r*, the real return associated with 10-Year Treasury securities.

The third figure above suggests that expected r* should return to the 4-quarter percent change in the potential level of the economy, 1.8 percent, from its current rate as of October 2017, 0.50 percent. Assuming that the inflation compensation portion of the rate on the 10-Year Treasury note returns to 2.0 percent, its currently estimated at 1.86 percent in October 2017, then the 10-Year Treasury note rate, which was 2.36 percent in October 2017, should reach 3.80 percent by 2019, (2.36 + 0.14 + 1.80 – 0.50).

However, because the pace of the run-off is expected to remain small relative to the size of the Fed’s balance sheet and the interest earned by the Fed’s balance sheet over and above the run-off caps will still be reinvested, then the net increase on the 10-Year Treasury note rate will be constrained. More precisely, the real return portion associated with the 10-Year Treasury note rate, the strongest channel by which balance sheet normalization will raise the rate on the 10-Year Treasury note, should rise, but is not expected to fully return to the 4-quarter percent change in potential GDP, 1.8 percent.

Recent forecasts produced by the Federal Reserve Bank of New York estimate that actual r*, which expected r* (the real return portion of the 10-Year Treasury note rate) has been shown to track, though over an abbreviated period, should reach 1.1 percent by the fourth quarter of 2019. Again assuming that inflation compensation returns to 2 percent, but also that expected r* continues to track estimates of actual r*, implies that the 10-Year Treasury note rate should reach 3.10 percent (2.36 + 0.14 + 1.1 – 0.50) by 2019.



Tags: , , , , ,

Leave a Reply

Your email address will not be published. Required fields are marked *