There’s Light at the End of the COVID-19 Tunnel
With recent vaccine developments progressing positively, an updated expectation regarding additional stimulus being passed, and stronger-than-expected incoming data, we upgraded our outlook for real gross domestic product (GDP). We now forecast 2020 full-year growth to be negative 2.2 percent (up from last month’s forecast of negative 2.5 percent) and 2021 growth to be 4.5 percent (up from 3.3 percent). Looking further ahead, we forecast 2022 growth to be a solid 3.2 percent, a two-tenths upgrade from November’s forecast. We continue to believe that the conditions for a continued, strong recovery are present once the limiting factors of COVID-19 on consumer behavior are lifted.
With that said, growth in the next few months is likely to be weak, and uncertainty remains high in the short run. The economy will have to navigate some obstacles before accelerating again. As of this writing, COVID-19 cases continue to rise, and hospitalization rates and daily fatalities are hitting record highs. In response, various state and local governments are increasingly implementing restrictions, which are likely to remain widespread with the advent of the holidays and potential greater exposures from large gatherings. Colder weather, independent of further distancing measures being implemented, also appears to be putting a damper on dining and other outdoor activities that experienced partial recoveries earlier in the year. It is also not clear how resilient holiday retail spending will be in the current environment.
Home sales continued their recent trend of remarkable strength, coming in stronger than we had anticipated in October. While robust housing activity is still expected over the next year, signs of a pullback from an extremely brisk pace are now appearing, and we expect sales in coming months to taper off somewhat. The current pace does not appear to be sustainable given the supply of listings and the current construction capacity of homebuilders. Still, we upgraded our total home sales forecast for Q4 to 7.60 million annualized units up from 7.16 million, and for 2021 to 6.66 million from 6.41 million. Housing starts were also upgraded for Q4 but adjusted modestly downward for 2021. Consistent with a faster pace of existing sales, as well as an upgrade to our expectations of refinancing activity, we increased our forecast for mortgage originations in 2020 by $169 billion to $4.29 trillion and by $745 billion in 2021 to $3.47 trillion. Originations for 2022 are forecast to be $2.84 trillion.
COVID-19 and Short-Run Weakness
While we upwardly revised our Q4 GDP forecast to 5.4 percent annualized (up from 4.3 percent), this pace of growth is almost entirely due to activity through the month of October. Real personal consumption expenditures (PCE) rose 0.5 percent over the month, somewhat higher than we had anticipated. This meant consumer spending entering November was at a level substantially higher than the average of the third quarter. If PCE were to remain flat month-over-month in November and December, this would translate into a quarterly annualized growth rate of 6.1 percent, slightly above our current Q4 forecast of 6.0 percent. Therefore, our forecast reflects an expectation of only slight growth in consumer spending for the month of November, followed by a modest decline in December. We further expect this weakness to continue into the first month or two of 2021 prior to a rebound occurring as spring approaches. Comparative weakness in labor market measures, namely a deceleration in employment growth in November to 245,000 jobs, down from 610,000 in October, as well as a recent bump up in unemployment claims, suggest that some softening is already underway. Auto sales also pulled back 3.1 percent in November, the first meaningful decline since April.
Risks in the near term remain high. It is a hopeful sign that COVID-19 cases in the upper Midwest states, which were generally the first to surge a couple months ago, have come down significantly over the last few weeks. However, the total national case count continues to rise. If other surging states do not soon follow the Midwest’s trend, we expect that more drastic consumer behavioral changes and state and local restrictions are likely to follow. We updated our assumption on the passing of an additional stimulus bill to not only include an extension of the broadened and long-run unemployment benefit programs, but also an additional $300 a week benefit. We continue to believe that, in the aggregate, households are in good shape to drive further consumption recovery even in the absence of additional fiscal support. However, as long as COVID-19 is limiting such behaviors, a sudden drop-off of unemployment benefits in January could have a substantial negative effect on consumer spending. There is a possibility that certain factors could cause first quarter growth to turn negative, for example, if new social distancing restrictions are widespread and an additional stimulus bill fails to pass. However, consumer behavior to date has held up remarkably well, and the rise in COVID case counts does not appear to have as strong of an inverse correlation with consumer spending as it once did. If strict lockdowns are avoided, and even modest consumer spending growth occurs in the near term, then the pace of recovery could exceed our baseline forecast.
Looking Beyond the Next Few Months
While uncertainty is elevated in the near term, the expectation of vaccine distribution commencing as early as the end of 2020 led us to project both a stronger recovery in consumer spending starting in the spring and a considerably faster pace of growth in the second half of 2021. Health officials have provided guidance that vaccines will be broadly available by April or so, but even before then, many higher-risk persons are likely to be vaccinated. We anticipate that a combination of factors will lead to an acceleration in consumer spending, including the vaccine rollout, reduced COVID-19 infection, hospitalization, and fatality rates, and warming weather leading to a return of outdoor activities.
The underlying conditions for strong growth appear to be present. Monetary and fiscal policies have been extremely accommodative, both within the U.S. and abroad. The financial system appears to be in much better shape than it was following the 2008 recession, with housing helping lead the recovery (an element missing in the wake of the Great Recession). Further, business inventory restocking is occurring at an accelerated pace compared to most cycles. Perhaps most importantly, consumers seem to be in a much stronger position to drive growth once COVID-19-related effects are no longer restricting behavior. In part due to unprecedented policy support, combined wages and salaries with unemployment benefits is up 2.0 percent from February, despite a sharp decline in the broader economy. In contrast, following the Great Recession, the equivalent measure was down 3.7 percent at a similar stage of recovery. Given the decline in consumption since March, this translates into a continued elevated savings rate. Even with the prior bonus $600 per week unemployment benefit expiring, the saving rate remained elevated at 13.6 percent in October, compared to a pre-March norm of about 7.0 to 7.5 percent. Consequently, consumers appear to have capacity to accelerate their spending from current levels.
The recoveries following the prior three recessions all included a substantial period of an increased saving rate, likely due to households seeking to repair their balance sheets in the wake of deflated asset bubbles. These periods of increased savings resulted in a drag on the pace of demand recovery. However, in prior recessions not characterized by asset bubbles, the saving rate tended to decline more quickly following the end of recession, and recoveries tended to be brisker. We believe this dynamic is more likely to occur in the present situation.
Since February, households have accumulated, in aggregate, about $1.6 trillion in additional savings above the pre-COVID trend. These additional savings, combined with rising equity and home prices, has led to a substantial increase in household net wealth. We believe the distribution of this additional wealth across households is relatively broad due to non-equity price drivers (e.g., home price gains, government support, retrenchment of consumption) being a large portion of the gains. At the end of Q3 2020, net wealth was 4.4 percent higher than in Q4 2019, a stark contrast to the experience of the Great Recession, during which net wealth, at its trough, fell 15.4 percent. We believe this provides a large amount of consumer spending power that was not present during the last cycle’s more anemic recovery.
Of course, risks remain that potential consumer spending does not begin by the spring of 2021. Uncertainty remains over the timing and breadth of vaccine distribution, both from a logistical standpoint and whether a large enough share of the population will be willing to participate. Another risk is over confidence due to vaccination efforts, which could lead to diminishing social distancing behavior prior to an adequate share of the population being immunized, resulting in additional infection waves and lockdowns. Beyond vaccine and COVID-related factors, it remains unknown to what extent permanent structural damage has occurred to varying business sectors and geographies, and if debt defaults and corporate bankruptcies will become a macroeconomic drag going forward. Certain industries will likely be slower to recover, and the extent to which capital and labor can quickly be reallocated to other purposes is uncertain. Additionally, while not anticipated in our baseline forecast, if growth does begin to accelerate, it is possible that inflationary expectations could also rise substantially, resulting in a significant jump in longer-term interest rates. Such an increase in rates could quickly dampen the housing and other sectors of the economy that have benefited from the current low rate environment.
Housing Strength Likely Will Continue, but Has to Cool Off
Existing home sales again surprised to the upside, rising 4.3 percent in October to 6.85 million annualized units, compared to an expectation of a modest decline. As a result, we have meaningfully upgraded our fourth quarter existing sales expectation to 6.62 million from 6.22 million (at an annualized rate), as well as increased the outlook into 2021 due to momentum from this year. While we still forecast a strong pace of sales in the first quarter of 2021, up 12.1 percent from the first quarter of 2020, we believe the sales pace will pull back from its October pace. Inventories continue to be extremely tight, with the months of supply of existing homes for sale hitting 2.5 in October, the lowest on record and down from 2.7 in September. Rapid home price appreciation has also eaten into the affordability gains from lower mortgage rates, with home prices up 7.3 percent in October from a year prior according to the CoreLogic National Home Price Index.
History suggests that the current sales pace for homes is not sustainable. October’s sales were above the normal relationship with the prior month’s pending sales index, which leads closings by 30-45 days. When these two series have diverged in the past, there is usually a convergence within a month or two. Further, the pending sales index dipped back 1.1 percent in October, and the Fannie Mae Home Purchase Sentiment Index® (HPSI) pulled back 1.7 points in November, both of which suggest a modest cooling of home sales in the near term.
Homebuyer demand remains strong, and we do not believe this is due entirely to low interest rates or remaining pent-up demand from the lockdowns in the spring. Fannie Mae Desktop Underwriter mortgage application data continues to show an elevated rate of urban dwellers purchasing homes in lower density areas in November. We do not know at this time whether this type of activity is a permanent shift in homebuyer preferences or a temporary phenomenon related to COVID-19 social distancing; however, for the time being, such activity appears to be further driving elevated home purchase demand. We expect this dynamic to be supportive of sales until at least the spring.
Low inventories of existing homes for sale combined with increased buyer demand for more suburban and exurban locations has also continued to support demand for new home construction. Single-family housing starts rose 6.4 percent in October to 1.18 million annualized units, the sixth consecutive month of gains and the highest rate since 2007. Homebuilders have been pushing to catch up with sales demand in recent months, as new home sales have eclipsed their usual relationship to starts, reflecting an abnormally high share of sales being fulfilled by homes not yet started. Therefore, we project that the current pace of new home sales is likely not sustainable, even as we expect home construction to be strong for the foreseeable future.
Though we foresee continued strength in home construction, we expect the pace of starts to pull back somewhat in coming months. The high current pace of home starts is in part attributable to fewer home starts in the spring. Due to these prior COVID-related disruptions, fewer homes than normal have been in the later stages of construction in recent months relative to pace of home starts. However, we believe we are at the point where total construction activity is likely approaching short-term homebuilder capacity constraints. Over the last decade, the seven-month rolling average of single-family starts, reflecting roughly how long it takes for the average home to be built, has correlated with measures of residential construction employment. The latter, representing builder capacity, moved up only modestly in recent months compared to the pace of starts, suggesting that the current starts pace will need to pull back to a more sustainable level represented by the trend in construction employment. We expect that construction employment will likely continue to expand going forward; however, it would have to be at a brisker pace than anticipated for the current pace of construction starts to be sustained in the near term.
Multifamily housing starts were flat in October, at 351,000. While this series is notoriously volatile, we continue to believe the underlying trend in multifamily construction will be weak for the foreseeable future relative to single-family construction. While suburban multifamily demand is expected to hold up better, the continued shift, at least for some renters, out of large metro areas and into homeownership, combined with continued remote working arrangements, will likely weaken demand for new projects in many areas. Until there is some clarity on what post-COVID-19 conditions will be, we believe urban-based projects in many areas are likely to be somewhat subdued.
For more on multifamily market conditions please see the December 2020 Multifamily Market Commentary.
Mortgage Originations Revised Upwards
The upward revisions we made to our purchase mortgage originations forecast this month were in line with the upgraded outlook for housing, as well as continued recent strength in incoming acquisition and securitization data. Purchase volumes are now expected to total $1.6 trillion in 2020, about $26 billion higher than last month’s forecast. Purchase volumes are expected to grow a further 7 percent in 2021 to nearly $1.7 trillion, an upgrade of $68 billion from the November forecast.
We expect refinance origination volumes in 2020 to reach $2.7 trillion, a $142 billion increase from last month’s forecast. Stronger-than-expected securitization and application data provide supporting evidence that this year’s refinance volume will surpass its historical peak in 2003. We expect this strong momentum in refinance activity to continue through 2021, given the continued low rate environment as well our estimate that nearly 70 percent of outstanding mortgage balances have at least a half-percentage point incentive to refinance. Our 2021 forecasted refinance volumes were revised upward significantly to $1.8 trillion, an elevated number compared to recent historical levels; however, this still represents a decline from 2020’s peak.
As mortgage rates edge closer to what we believe will be the trough at 2.7 percent, and the 10-year Treasury yield has remained at or above 90 basis points so far in December, mortgage spreads continue to compress. While there may be some resistance to the 10-year yield consistently pushing above 1.0 percent in the coming months, with vaccination efforts and subsequent stronger economic growth, we expect Treasury yields to move higher over the next year.
The primary spread (30-year mortgage rate minus the 10-year Treasury yield) in early December fell to approximately 180 basis points, a level similar to those seen at the beginning of the year. The compression in the spread was consistent with Fannie Mae’s Mortgage Lender Sentiment Survey® (MLSS) for the fourth quarter, which showed lenders’ profit expectations, on net, declined from the previous quarter. Furthermore, lenders are expecting demand growth for both purchases and refinances over the next three months to slow, which aligns with our forecast of slowing home sales heading into 2021.
Economic & Strategic Research (ESR) Group
December 12, 2020
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: Department of Labor, Federal Reserve, National Association of REALTORS®, 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.
ESR Macroeconomic Forecast Team
- Doug Duncan, SVP and Chief Economist
- Mark Palim, VP and Deputy Chief Economist
- Eric Brescia, Economist
- Nick Embrey, Economist
- Rebecca Meeker, Financial Economist
- Richard Goyette, Business Analyst