View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

ON THE ECONOMY | DECEMBER 2019
https://www.stlouisfed.org/on-the-economy/2019/december/recession-signals-housing-indicators-remain-consistent-broaderslowdown-2020

Recession Signals: Housing Indicators Remain
Consistent With a Broader Slowdown in 2020
By William R. Emmons

KEY TAKEAWAYS
Over the past year, four housing indicators have moved in
ways consistent with patterns seen before three previous
recessions.
These indicators are mortgage rates, existing home sales,
real house prices and the momentum of residential
investment.
More recent housing data still point to a slowdown, albeit a
less severe one.

A year ago I noted that recent movements in four key housing indicators—mortgage rates, existing home
sales, real house prices and the momentum of residential investment—resembled those seen in the late
stages of past economic expansions.1 The trends for all four indicators as of the third quarter of 2018 lined up
well with previous cyclical patterns when projecting a recession start date in the fourth quarter of 2019.
Moving this onset date a few quarters in either direction did not change the basic implication that housing
developments at that time were signaling a possible near-term recession.
One year later, three of these housing indicators have changed very little, while one—the 30-year mortgage
rate—has dropped significantly. Still, all four indicators remain consistent with a recession beginning in late
2019 or early 2020 based on their behaviors in previous business cycles.

Year-Ago Outlook Holds with Latest Data
The most important insight provided by four additional quarters of data is that the current business cycle
appears more similar to the periods before the relatively mild 1990-91 and 2001 recessions than the very
severe 2007-09 Great Recession, at least in terms of housing indicators. There is little evidence from the
more recent data that recession risks have disappeared.
In the following sections, I update the four housing indicators through the third quarter of 2019 to gauge the
signals they may be sending about a potential near-term recession.

1. 30-Year Fixed Mortgage Rates

Consistent with earlier cycles, the average 30-year fixed mortgage rate has declined significantly even though
a recession has not begun. Figure 1 updates the path of this long-term mortgage rate through the third
quarter of 2019.2 The last observation of about one-half of one percentage point below the most recent threeyear moving average is identical to the level one quarter before the onset of the 2001 recession and is just
below the level one quarter before the 1990-91 recession. Hence, this indicator remains consistent with—
though still does not guarantee—an imminent recession.

Figure 1

SOURCE: Freddie Mac.
NOTES: The figure shows the average quarterly 30-year fixed mortgage rate minus the average of the
previous three years (12 quarters). Each line shows three years of data before and after a recession; time
zero is the quarter in which a recession began. The first quarter of the recession is indicated in the line label
(key): the fourth quarter of 1990, second quarter of 2001 and first quarter of 2008.
*This line assigns the beginning of the next recession (period zero) to the fourth quarter of 2019; of course,
the onset of the next recession is unknown. The last observation in this series (for quarter -1) is for the third
quarter of 2019.

2. Existing Home Sales

The pace of existing home sales relative to their recent trend rate continued to slow during 2019. (See Figure
2.) Despite a modest upturn in the third quarter of 2019, this indicator remains firmly in the range observed
prior to the 1990-91 and 2001 recessions. In contrast to the Great Recession, the decline in home sales since
dipping below zero has been much more moderate.

Figure 2

SOURCE: National Association of Realtors.
NOTES: The figure shows the percent difference between the current rate of existing single-family home
sales (four-quarter average) and the average annualized sales rate during the previous three years (12
quarters). Each line shows three years of data before and after a recession; time zero is the quarter in which
a recession began. The first quarter of the recession is indicated in the line label (key): the fourth quarter of
1990, second quarter of 2001 and first quarter of 2008.
*This line assigns the beginning of the next recession (period zero) to the fourth quarter of 2019; of course,
the onset of the next recession is unknown. The last observation in this series (for quarter -1) is for the third
quarter of 2019.

3. Real House Prices

Qualitatively, the recent behavior of inflation-adjusted home-price growth relative to its recent trend rate
(shown in Figure 3) is very similar to the patterns of the two previous indicators—that is, it mirrors the run-up
to the relatively mild 1990-91 and 2001 recessions and is unlike the pattern just before the Great Recession.

Figure 3

SOURCES: S&P CoreLogic and U.S. Bureau of Economic Analysis.
NOTES: The figure shows the four-quarter percent change in the CoreLogic Home Price Index, deflated by
the Personal Consumption Expenditures (PCE) Chain-Weighted Price Index, minus the annualized percent
change during the previous three years (12 quarters). Each line shows three years of data before and after a
recession; time zero is the quarter in which a recession began. The first quarter of the recession is indicated
in the line label (key): the fourth quarter of 1990, second quarter of 2001 and first quarter of 2008.
*This line assigns the beginning of the next recession (period zero) to the fourth quarter of 2019; of course,
the onset of the next recession is unknown. The last observation in this series (for quarter -1) is for the third
quarter of 2019.

4. Contribution of Residential Investment to GDP Growth
If any of the highlighted housing indicators hints at a departure from the typical recessionary pattern it is the
contribution of residential investment to GDP growth. (See Figure 4.) Despite some improvement over the
course of 2019, residential investment continued to be a drag on economic growth through the third quarter,
albeit a slightly smaller one than before. Of all four indicators, this one most closely resembles the patterns
seen before the two relatively mild recessions of 1990-91 and 2001 versus the more severe Great Recession.

Figure 4

SOURCE: U.S. Bureau of Economic Analysis.
NOTES: The figure shows the four-quarter average contribution to real GDP growth minus the annualized
contribution during the previous three years (12 quarters). Each line shows three years of data before and
after a recession; time zero is the quarter in which a recession began. The first quarter of the recession is
indicated in the line label (key): the fourth quarter of 1990, second quarter of 2001 and first quarter of 2008.
*This line assigns the beginning of the next recession (period zero) to the fourth quarter of 2019; of course,
the onset of the next recession is unknown. The last observation in this series (for quarter -1) is for the third
quarter of 2019.

Conclusion: Housing Indicators Still Signal Recession, Albeit a
Less Severe One

The value of leading indicators—from housing variables to the slope of the yield curve—is that they offer an
opportunity to prepare for a possible economic slowdown or outright downturn. A recent example is the
Federal Reserve’s dramatic turn from a program of monetary tightening in 2018 to an easing of policy in
2019.
One hazard of leading indicators is that they can lead to misinterpretation and complacency. If the economic
slowdown signaled by weakening housing indicators and some portions of the U.S. Treasury yield curve
inverting in 2019 does not begin immediately, some observers may think the precautions undertaken in
response to the signals, such as the Fed’s recent easing of monetary policy, can void the signals themselves
and pre-empt a recession.
This would run counter to the historical patterns documented in this article: The Fed eased monetary policy
and mortgage rates plunged in advance of each of the three previous recessions, yet the economy still went
into a downturn. Thus, the value of leading indicators may lie more in their role as early warning signals that
help us better prepare for, rather than outright prevent, a recession.
This time could be different, however, if the Fed’s timely interest rate cuts and other factors in fact help to
prevent a recession in late 2019 or 2020. If that happens, we should re-examine the indicators that have
been successful in signaling recessions in the past. In the meantime, we should not dismiss their salience.

Endnotes

1. See Emmons, William R. “Recession Signals: Four Housing Indicators to Watch in 2019.” Housing Market
Perspectives, December 2018.
2. The placement of the respective housing indicator in this and all subsequent figures assumes that a
recession will begin in the fourth quarter of 2019, even though the beginning date of any future recession is
unknown. I make this assumption simply to place the current cycle in a plausible position relative to previous
cycles.

This article originally appeared in our Housing Market Perspectives publication.