Supply Constraints Increasingly Come into Focus as GDP Approaches New High

Our 2021 real gross domestic product (GDP) forecast was modestly revised upward to 7.0 percent from a prior 6.8 percent on a fourth quarter-over-fourth quarter basis. This faster pace was driven by stronger first quarter growth than we expected, combined with a slightly improved outlook for near-term consumer spending. We believe the greater consumer expenditures over the first half represent a pulling forward of some growth from the second half of 2021 and early 2022, compared to our previous forecast. Our 2022 growth forecast was revised downward to 2.8 percent from a prior 3.0 percent. Recent indicators point to a continued recovery in the near term as COVID-19-related restrictions wane and the vaccination effort progresses. With second quarter growth forecast of 9.2 percent annualized, we expect this quarter’s GDP level to surpass the pre-COVID peak. However, supply chain disruptions, labor scarcity, and inflationary pressures are increasing risks to future growth. While we had previously expected the rate of inflation to modestly accelerate into 2022, we have revised upward our inflation forecast and now expect that the annual change will not fall below the Federal Reserve’s long-run 2.0 percent target within the forecast time horizon.

We downgraded our second quarter existing home sales forecast, but this was roughly offset by an upgrade to expected sales later in the year, in addition to a modest upgrade to new home sales. Combined, forecast revisions to total annual sales were minimal. We now expect 2021 sales to increase 6.3 percent for the year, compared to 6.2 percent previously. While demand for home purchases remains robust during the current spring buying season, a lack of listings is limiting the pace of purchases and putting upward pressure on home prices. This lack of existing homes for sale is helping bolster demand for new home construction. However, homebuilders are facing their own supply constraints, most notably the price of lumber and other materials, as well as a lack of buildable lots and hiring difficulties. While we forecast single-family housing starts to be 24.8 percent higher in 2021 compared to 2020, the construction pace would likely be even greater if not for the previously mentioned limiting factors. Our mortgage originations forecast was little changed. A slight downward revision to purchase originations based on updated incoming data was more than offset by an upgrade to refinancing activity on a modestly lower interest rate outlook. We forecast 2021 single-family mortgage market originations in 2021 will be $4.1 trillion, up from $4.0 trillion. Our forecast for 2022 originations remains at $3.0 trillion.

Supply Side Risks Becoming More Prevalent
In prior commentaries, we routinely highlighted risks around continued consumer spending recovery. Uncertainty over how extensively consumers will tap into roughly $3.5 trillion of accumulated savings from the past year and to what degree they will hesitate to return to previously restricted activities represent considerable risks to our GDP forecast. While this uncertainty, along with unknowns about future vaccination efforts, is still high, recent developments highlighted the growing comparative risk around constraints to the productive side of the economy. Surging commodity prices, continued supply chain disruptions for inputs, including semiconductors, and possible labor market scarcity represent downside risks to growth and upside risks to inflationary pressure. If surging demand intersects with a tighter-than-expected supply side of the economy, faster inflation and higher interest rates are likely. This may lead to an earlier-than-expected tightening of monetary policy, which could weigh on both growth and asset prices, and become a meaningful drag on home sales, depending on the magnitude of the increase in rates.  

Beyond macroeconomic uncertainty, the housing outlook faces additional risks. We continue to believe that the bulk of abnormally high home purchase demand over this last year was due to a combination of last year’s delayed homebuying season and households’ pulling forward their plans to move. Additionally, stimulus checks have aided recent homebuyers’ ability to make down payments. Thus, our housing forecast is, in part, motivated by a view that as these factors dissipate, homebuyer demand will likely begin to wane later in the year. However, if people increasingly reassess their desired housing arrangements post-COVID, then a heightened level of sales could potentially be sustained for a longer period. Additional uncertainty involves the timing and implications of an end to forbearance policies and how that might impact the number and nature of home sales. However, we believe continued improvement in the labor market and higher levels of home equity will likely help limit distressed sales.   

Employment Gains Soften; Signs of Hiring Difficulty Grow

The April employment report from the Bureau of Labor Statistics was disappointing. Market expectations were for a gain of near a million jobs, but the figure reported was a much smaller 266,000. We think that this comparatively weak reading was likely, in part, an anomaly. A combination of volatility in the recovery path, sampling issues with firm openings and closures, and atypical seasonal patterns may have led to the weaker number. The alternate ADP measure of private sector payroll employment showed a much stronger 742,000 jobs gain over the month, and a continued acceleration trend. The swift decline in initial unemployment claims over the period also suggests a much stronger pace of improvement. If this weak labor report is indicative of a faltering recovery, little evidence from other indicators support that claim. Auto sales, an early reading of consumer activity, rose in April to the highest level since 2005, following a surge in March. This occurred despite continued chip shortages dragging on auto manufacturing. High frequency measures, such as air travel passenger counts, restaurant reservations, and debit and credit card spending, all suggest continued expansion of activity as COVID-19-related restrictions wane. Consumer confidence measures also jumped in April, and, while the ISM manufacturing survey pulled back, it sat at 60.7, a level consistent with strong manufacturing sector growth. While the measure is delayed an additional month, job openings moved up again in March, hitting a record high of 8.1 million. These indicators suggest robust labor demand, despite the weak April jobs number, lending some support to the theory that this most recent reading was an anomaly.  

However, further details within the employment report are consistent with a growing list of anecdotes from employers stating difficulties in finding workers. Job gains are perhaps being held back by supply-side factors. The average hours worked per week rose by one-tenth, and the share of workers working part time who would like to work full time fell sharply – both consistent with growing labor demand and employers having difficulty finding new workers. Additionally, despite the bulk of job gains over the month occurring in lower-wage sectors, the month-over-month average hourly wage rate rose by a brisk 0.7 percent, suggesting employers began to bid up wages to attract employees. Additionally, the quits rate moved up in March to a level similar to the pre-COVID rate. Quits tend to rise when workers feel confident enough to find other work. The quits rate therefore indicates a tighter labor market than the unemployment rate suggests. The NFIB survey showed in April a record share of small businesses reporting job openings unable to be filled, and “quality of labor” as their single most important issue. While it is uncertain how much this labor scarcity dynamic was the driving force behind the disappointing April employment report, this is clearly a growing drag.

We believe these tight labor supply characteristics are likely occurring despite total employment still being about 8.2 million below the pre-COVID peak due to a mix of factors: 1) there is still a segment of workers who either cannot or do not wish to return to work due to COVID-19 worries; 2) school closures and a related surge in childcare prices are keeping some workers from reentering the workforce; 3) the relaxing of job search requirements in qualifying for unemployment benefits and the expanded size of the benefit may be keeping many unemployed from looking for employment; and 4) there is likely a significant geographic and industry mismatch between where labor is demanded and where workers are available. These effects should wane as COVID-19 concerns diminish, schools reopen, and expanded unemployment benefits expire. Yet, if this mix is currently restricting employment growth to the extent that many anecdotal stories suggest, rather than the last report being a data anomaly, labor scarcity will likely weigh heavily on economic growth. For now, our forecast implies a rebound in employment gains back to about 900,000 in May. However, if that report ends up disappointing as well, then our current near-term growth outlook would likely be revised meaningfully downward.

Inflation Comes in Hot
Following the completion of our forecast, but prior to this writing, the April consumer price index (CPI) report showed a greater-than-anticipated rate of inflation. The headline CPI jumped to 4.2 percent on an annual basis, up from 2.6 percent a month earlier. The core index, which removes volatile food and energy prices, rose to 3.0 percent, up from 1.6 percent in March. For headline CPI, this was the fastest annual rate of growth since September 2008, while for core CPI it was the fastest annual rate of growth since January 1996. While some of this annual gain was due to base effects (where the annual comparisons reflect the brief deflationary period a year prior following the initial COVID-19 lockdowns), the more recent trend shows acceleration. On a monthly basis, the headline and core indices rose 0.8 and 0.9 percent, respectively, the largest monthly increases since June 2009 and April 1982.  

There is reason to believe that much of this jump in prices is due to one-off readjustments and temporary factors. Over a third of the monthly rise in headline CPI was driven by a 10.0 percent surge in the price of used cars, reflecting ongoing supply chain problems in auto manufacturing, as well as other COVID-19-related auto market dynamics. Additionally, much of the remaining rise was driven by sectors related to reopening now experiencing pricing power for the first time in a year. These include categories such as airline tickets, hotels, and eating establishments. Backing out auto purchases and these “reopening” sectors, the CPI would have risen by a lesser 0.4 percent over the month, a level in line with our forecast. We believe the headline number therefore exaggerates the underlying inflationary trend. Still, pricing pressures are clearly growing, and we continue to see upside risk to inflation going forward. Ongoing monthly gains of 0.3 to 0.4 percent translate into an annualized pace of inflation between 3.5 and 5.0 percent, a level well above the Federal Reserve’s long run 2.0 percent target.

For now, we believe the Fed is likely to continue to see the recent jump in inflation as a “transitory” increase, and therefore is unlikely to change its guidance. Following the inflation report, Fed vice chairman Richard Clarida downplayed it as “one data point.” Market participants seemed to largely discount the report as well, with no fundamental shift in monetary policy expectations. The Fed will get to see both another monthly inflation reading and employment report prior to its June meeting, where it will once again have to decide whether to provide additional clarity regarding the criteria or timing for reducing its purchases of Treasury and mortgage-backed securities. While our baseline forecast expects much of this inflation jump to pass, we still expect a modestly accelerating underlying inflation trend and see considerable upside risk going forward. Prior to the release, our forecast was for core CPI to remain above 2.0 percent annualized over the forecast horizon and to finish 2022 at 2.7 percent.  

Supply Constraints Are Also Limiting Housing Activity
As expected, existing home sales fell in March, declining 3.7 percent to an annualized pace of 6.0 million units. While the limited supply of homes available for sale continued to drag on sales, the month’s decline was in part driven by the cold weather and related power outages a month earlier (existing sales are measured at the point of closing, which typically occurs 30-45 days after signing). While the decline was expected, recent data suggest that the forecast rebound in April sales will be smaller than we had previously anticipated. Pending sales, which measure contract signings, moved only modestly upward in March, rising 1.9 percent. Purchase mortgage applications have also trended lower in recent weeks. As such, we revised downward our forecast for second quarter existing sales to 5.88 million from 6.16 million annualized units. We have long expected that a combination of waning COVID-19-induced movement into single-family housing and continued tight inventories would lead to a slowing pace of existing home sales as the year progresses. However, the latter factor appears increasingly more limiting. The months’ supply of homes for sale at the end of March, even with the sales pace slowdown, was only 2.1 months, a near record low. More timely weekly data published by Redfin also indicate a record-high 48 percent of homes sold above list price in April and a record 45 percent of pending home sales under contract within 7 days of listing. For comparison, these metrics for the 2019 spring buying season were around 24 percent and 28 percent, respectively.

With demand continuing to outstrip the supply of listings available, home prices continue to surge. As measured by the CoreLogic National Home Price Index, home prices were up 11.3 percent in March from a year prior, the highest annual growth since 2006. While low interest rates and elevated savings from the past year are allowing buyers to absorb rising prices for now, continued strong price appreciation will increasingly weigh on affordability. The Fannie Mae Home Purchase Sentiment Index® (HPSI) pulled back in April by 2.7 points, driven in large part by a decline in the share of respondents indicating it was a good time to buy. This measure in turn was dragged on by a growing share of respondents pointing to elevated home prices and a lack of homes available for sale.

Strong house price appreciation and a lack of available existing homes for sale continued to bolster demand for new homes. New home sales jumped 20.7 percent in March to an annualized pace of over 1 million units, the fastest pace since August 2006 (new sales are recorded at the point of contract signing). Single-family housing starts also rose over the month, increasing 15.3 percent. We modestly upgraded our single-family starts forecast, and we now expect an increase of 24.8 percent in 2021 compared to 2020. Homebuilders would probably increase the pace of construction even more if not for the supply constraints they currently face. Lumber prices continue to ascend, now up over 300 percent from February 2020 levels according to the NASDAQ. Meanwhile, a modest residential construction employment gain in April was consistent with difficulty hiring additional labor. The total level of homes for sale, regardless of the stage of construction, has largely moved sideways over the past year even as sales grew swiftly. Homebuilders are not only limited by material and labor costs, but also by lot availability. Until some of these supply constraints can be alleviated, a further acceleration in starts will likely be limited, even as demand for new home construction remains robust.

Multifamily starts also rose in March, rebounding 30.8 percent to an annualized pace of 501,000 units, following a weather-induced decline of 21.8 percent in February. Weather aside, multifamily construction has been surprisingly resilient in recent months, as demand for units in many smaller and less expensive metro areas and suburban locations likely drove developer interest. With rent metrics now moving upward and vacancy rates remaining low, we expect a healthy pace of multifamily construction moving forward. Still, the core of some larger metro areas will likely see continued weaker activity in the near term. We made slight upward changes to our multifamily construction forecast and now project starts in 2021 to be 410,000, 5.5 percent above 2020.

For more on multifamily market conditions please see the May 2021 Multifamily Market Commentary.

Interest Rates Take a Breather after Rapid Rise to Start the Year
Interest rates remained in a relatively tight band throughout April, with the 10-year Treasury starting the month at 1.69 percent and ending the month at 1.65 percent. The month averaged 1.64 percent, a modest 3 basis point increase from March but the highest level since January 2020. The pace of growth in rates has slowed considerably from the rapid rise earlier this year.

Mortgage rates have fallen somewhat since their recent peak of 3.18 percent the week of April 1. The week of May 6, the 30-year fixed mortgage rate sat at 2.96 percent, according to Freddie Mac. This was the third consecutive week of the mortgage rate sitting below three percent. Given the decline in mortgage rates and the slight increase in the 10-year Treasury, mortgage spreads compressed over the month, with the primary spread (30-year mortgage contract rate minus the 10-year Treasury yield) narrowing for the twelfth consecutive month to 143 basis points, the smallest gap since April 2011.

This month we made relatively modest revisions to our forecast for purchase mortgage originations. We downgraded our expectation for 2021 purchase volumes by $43 billion from last month’s forecast to $1.8 trillion. Our forecast for 2022 purchase volumes remains at $1.9 trillion, essentially unchanged from last month.

We expect refinance origination volume to be $2.2 trillion in 2021, a $125 billion upward revision from last month’s forecast, as incoming data continue to come in strong and interest rates have pulled back in recent weeks. Our lower forecasted interest rate path suggests there could be steam remaining in the current refi boom. We expect refinance volume in 2022 to total $1.1 trillion, an upward revision of $43 billion from last month’s forecast, but a decline of 49 percent from 2021. We will continue to monitor our expectation around interest rates and the effects of future rate volatility. At current interest rates, we estimate around 51 percent of all outstanding mortgages have at least a 50-basis point incentive to refinance, up from 42 percent in last month’s forecast given the recent rate declines.   

Economic & Strategic Research (ESR) Group
May 12, 2021
For a snapshot of macroeconomic and housing data between the monthly forecasts, please read ESR’s Economic and Housing Weekly Notes.

Data sources for charts: Automatic Data Processing Inc., Bureau of Labor Statistics, Census Bureau, Fannie Mae ESR Group.

Opinions, analyses, estimates, forecasts and other views of Fannie Mae's Economic & Strategic Research (ESR) Group included in these materials should not be construed as indicating Fannie Mae's business prospects or expected results, are based on a number of assumptions, and are subject to change without notice. How this information affects Fannie Mae will depend on many factors. Although the ESR group bases its opinions, analyses, estimates, forecasts and other views on information it considers reliable, it does not guarantee that the information provided in these materials is accurate, current or suitable for any particular purpose. Changes in the assumptions or the information underlying these views could produce materially different results. The analyses, opinions, estimates, forecasts and other views published by the ESR group represent the views of that group as of the date indicated and do not necessarily represent the views of Fannie Mae or its management.