- Even after the immediate threat of COVID has receded, we expect as many as 50% of office-based employees will work from home at least once a week. But the move away from cities and toward the suburbs should prove short-lived. Most of those leaving cities will have brought forward a planned move, and changes in relative prices will rebalance demand. With more time spent at home, we expect a rise in overall housing demand. However, given income constraints, the impact on house prices in the long run will be modest. In the US, increased demand for larger apartments, coupled with a rise in supply from empty office buildings, implies rents and capital values will see small falls from 2025 to 2029.
- Home working has surged in response to the coronavirus. Looking ahead, even after the threat from the virus recedes, we expect perhaps 50% of office workers will continue to work remotely in some form. That could have two impacts on the residential market. It may alter where people choose to live and change the type of accommodation demanded.
- In terms of location, increased working from home will mean lower commuting costs. That will make living further from transit stations and city centres more palatable, and we expect this to trigger a modest shift in demand towards suburban versus city centre locations. But most of that change in demand will reflect households bringing forward a planned move to the suburbs. The arrival of the coronavirus has convinced a bulge of young workers in their late-20s that now is a good time to move.
- That means house prices in suburban locations will outperform, but only briefly. Once those who were planning on moving have done so, the exodus from cities will run out of steam. After all, cities have a lot more to offer than just a shortened commute and, with suburban house prices rising in the near term, city centres will start to look relatively cheap, balancing out demand. In the long run, the virus will not drive an exodus from cities any more than the high level of house prices already has.
- With people spending more time at home, we expect an increase in demand for larger and better quality homes and apartments. That will put upward pressure on house prices, but any impact will be limited by existing affordability constraints. Shifting demand preferences will also trigger a supply response. In the suburbs, where land is more readily available, increased supply will help to dampen price rises. Once the immediate impact of the virus has passed, we therefore do not expect a significant impact on house prices, which will rise in line with earnings growth in both the US and UK.
- Looking at the rental market, working from home one or two days a week will increase demand for larger units with better amenities. The good news is that, with plenty of empty office space available, supply should respond and apartments will start to get larger for the first time since 2006. The bad news is that with rents also constrained by incomes, and with yields set to remain close to 4%, rental growth and capital values are set for a prolonged period of subdued or negative growth. We except apartment capital value growth in the US to average -0.4% y/y from 2025 to 2029.
- This publication follows the first in our Future of Property series, which considers the long-term outlook for real estate. To learn more about the series, please contact your account manager or sales executive.
The outlook for housing in a post-COVID world
Our series on the future of property examines how the coronavirus will impact property markets in the long run. By long-run we mean a time when the coronavirus is no longer a threat, for example due to the development of a vaccine.
In our previous Focus, we concluded that even after the coronavirus pandemic subsides, part of the rise in working from home will be sustained. In turn this would limit growth in demand for office space – lowering rental values across the next decade.
In this Focus, we examine how the virus will impact housing markets in the US and UK. In particular, we look at how the move towards working from home will affect where households choose to live, the types of home or apartments demanded, and what that will mean for house prices and rents over the next decade.
Home working has surged due to COVID
In our first Focus in the Future of Property series, available here, we explored the outlook for home working in detail. In summary, pre-COVID, a relatively low proportion of office workers spent most or all their time working remotely, even though the technology to do so had been in place for some time. For example, in the UK, Eurostat data show that 4.7% of employed persons reported home as their main place of work in 2019. In the European Union, the same figure stood at 5.3%.
The immediate impact of the virus changed this. European surveys show a dramatic shift in the share of those working from home. This shift was particularly pronounced for “white-collar” jobs, which is a good proxy for office-based roles. Most of such workers in many of Europe’s biggest economies worked from home on a regular basis in the spring and summer months. (See Chart 1.)
Chart 1: Survey of White-Collar Workers (June 2020, % of Respondents)
Sources: Cass Business School, IESE Business School, SD Worx
Part of this shift will become permanent
Prior to the coronavirus, the effects of extended home working were not well understood by most businesses. But due to the coronavirus response, the last six months have been a largely successful live experiment in home working. This seems to have broken down many of the barriers to a more remote workforce, as many who were formerly doubtful now believe that at least some degree of remote working could be a net benefit to both their business and their employees. As a result, after the direct effects of the pandemic subside, we expect to see a permanent shift toward more working from home.
In the US offices sector, around 60-70% currently work from home. As the risk from the pandemic retreats and working practices move towards a new normal, the share will fall back from this very high level. But, even by the end of 2022, we think it is possible that a total of 50% of workers will be working remotely in some form, with a little over 20% working at home full-time and 30% doing so for at least one day per week. (See Chart 2.)
Chart 2: Share of US Office-Based Workers Working At Home (%)
Sources: US Census, Capital Economics
Beyond that, we expect a slow rise in those working at home one day a week over the remainder of the decade, though we think that 20% could be the ceiling for those working at home full-time, at least for the next decade. This reflects the benefits of the office environment, although we acknowledge there is huge uncertainty around this split.
Outside of offices, fewer jobs can be done remotely. Hospitality, manufacturing and agricultural jobs won’t be done from home after the pandemic ends. And across the US, Europe and UK, the share of office-based employment hovers at around 20% to 25% of the workforce. That suggests that, looking at the workforce as a whole, a majority will not see an increase in more home working. (See Chart 3.)
Chart 3: Office-based Employment (% of Workforce)
Sources: Refinitiv, Capital Economics
Urban productivity premiums will largely hold up
The disproportionately large impact on office-based jobs suggests that residential markets in cities – where office work dominates – will be most affected by the shift to home working.
A key driver of house prices in cities is the urban “productivity premium”. This premium reflects the fact that cities tend to have clusters of jobs in higher productivity occupations which offer their workers higher pay. Living in a city with those opportunities raises a workers’ earnings, which in turn boosts purchasing power. With the supply of homes limited due to land constraints, competition for homes in cities will bid up prices and rents.
This can be seen in large global cities such as London, Paris, New York or San Francisco. Houses there have tended to be more expensive than homes elsewhere in part due to higher average wages. In the UK for example, London house prices and earnings are a clear outlier – with earnings partly explaining the high level of prices. (See Chart 4.) But other regions with higher wages also tend to have higher house prices.
Chart 4: House Prices and Wages by UK Region
Sources: Refinitiv, Capital Economics
If more office-based employees work from home, they may no longer need to live in a city with a high productivity premium. That will act to reduce demand and price pressures in cities, while raising them where workers choose to relocate.
That said, the extent to which the productivity premium will be affected depends on how detached home workers can become from their work location. As we have mentioned previously, we expect most office workers will still need to access a central office occasionally. That will keep high-paid employees in or near cities.
Furthermore, the productivity premium extends beyond just the workplace. Cities also provide access to professional networks and social opportunities. Even those working remotely will still value access to those benefits.
Urban amenity premiums will fall
Beyond the productivity premium, housing is also valued based on the availability and quality of local amenities. After all, the house price-to-earnings ratio in London is higher than other regions, showing that higher earnings on their own cannot fully explain London house prices.
For example, cities offer an increased choice of entertainment and restaurants, whereas suburban locations provide more green space and reduced crime. Transport is a key amenity in all areas, and the value that households attach to transport access is likely to change as working from home increases.
Currently, households value proximity to transport because it can reduce their day-to-day travel cost. This includes the direct cost of travel, as well as the time cost. Data for the US show the average worker spends $2,600 a year travelling to work. On top of that, transport time is also significant. The average commute time was 65 minutes per day in the UK.
Proximity to transport amenities are therefore an important driver of house prices. In the UK, data published by the Nationwide for 2019 show that, in London, buyers are willing to pay an additional 9.4% for a house within 500m of a subway compared to a house 1,500m away. (See Chart 5.)
Chart 5: House Price Premium by Distance from Rail Connection (%, compared to 1,500m)
Sources: Refinitiv, Capital Economics
The premium for living near mass transit is smaller in other cities but still significant. For example, the premium for Manchester is 7.8% for a home 500m away from a local transit station. By reducing the need to travel, an increase in working from home will reduce demand for housing near public transport hubs – lowering the associated price premium on housing.
That said, any downward effect on prices shouldn’t be overestimated. Well-connected homes in large cities might attract a 10% premium, but that premium is not just for transport. Transport links are often located on local high streets which contain other amenities such as bars, restaurants and retail. Furthermore, transport links are also used to access recreational amenities elsewhere.
Disentangling the premiums on transport, work and recreation proximity is tricky. But it is clear that proximity to recreational amenities adds value to homes beyond the advantage of a quicker commute.
A short-term move from cities to suburbs
Overall, once the pandemic has passed, we do not expect the coronavirus to trigger a major adjustment in cities’ productivity or amenity premium on housing. After all, the majority of office workers will still need to go into the workplace sometime during the week. The most expensive global cities such as London and New York continue to confer significant benefits to those who live there – in terms of both a productivity premium but also in terms of amenities.
Housing demand in such cities will therefore not see a major downward adjustment in the aftermath of the pandemic. That said, demographic factors mean a short-term shift from cities to the suburbs is certainly a possibility. After all, such a trend has already been happening in the US for the last few years. Net internal migration in the largest metros, with the exception of a few in the South, has been negative for some years now. For example, internal emigration from New York City rose steadily over the last decade to 140,000 by 2017. (See Chart 6.) Similarly, there is evidence of a surge in internal migration from London.
Chart 6: New York City Migration (000s)
Sources: Census Bureau, Capital Economics
Migration from cities looks to, in large part, reflect demographics. A bulge of Americans reached their late 20s/early 30s over the past couple of years. (See Chart 7.)
Chart 7: US Population by Age (Millions)
Source: Census Bureau
That is an age when many decide to leave cities, as the benefits of greater access to city amenities begin to fall, and households look to buy cheaper, larger homes to start families. Indeed, data from the UK shows that while younger adults move to cities on a net basis during their 20s, people start to leave London for other parts of the country from around the age of 30. (See Chart 8.)
Chart 8: London Net Internal Migration
Sources: Refinitiv, Capital Economics
The outlook for demand for homes and apartments in cities will therefore also depend on demographics. Admittedly, the demographic bump in the US is not as pronounced in the UK, where other factors are more significant. Notably, the rise in migration away from London has probably been driven by the sharp deterioration in housing affordability over the last decade, combined with its wide house price differential with the rest of the UK.
In either case, in the short term, concerns over the coronavirus and a shift to increased working from home may encourage some of those in their late-20s to bring forward a planned move to a larger home. That ‘cohort’ effect will boost migration out of cities over the next year or so.
But beyond the short term, outward internal migration from cities is likely to drop back. For a start, in the US, there is no bulge of American’s currently in their early-to-mid-20’s. (See Chart 7 again.) Without the outward migration of those households over the next decade, that should help to offset any drop in city demand due to increased working from home.
The prospects for international migration are also key. Large cities are magnets for international migrants, both for the employment opportunities they provide and local migrant communities. Indeed, US and UK city populations have been supported by international migration over the past few years. In NYC, net international migration has averaged 48,000 per year over the past five years. (See Chart 6 again.)
Fears about the coronavirus, uncertainty and quarantines have sharply cut international migration flows in recent months. But there has always be strong demand from migrants to move to large cities, and we doubt that will change once the virus is no longer an issue. Indeed, if governments start to worry that cities are emptying out, they may decide to allow or encourage more migration.
Suburban house prices to outperform in short-term
Overall, in the US once the initial cohort effect of Americans in their late-20s has worked its way through, we expect city populations will stabilise. And we expect a similar pattern in the UK. But, in the short term, a flight from cities will help support housing demand and house prices in the suburbs.
Indeed, that already appears to be happening. In the UK, data from property website Zoopla show that the drop in the time it takes to sell a single-family home in 2020 compared to 2019 has been larger compared to multifamily properties. (See Chart 9.) This in part reflects tax changes, which have temporarily made larger homes cheaper to buy. But a desire for a home in the suburbs, rather than an apartment in a city, is likely to be another underlying factor driving this trend.
Chart 9: UK Time to Sell (Days)
And, in the US, there has been a sharp rise in the single-family home share within total existing home sales over the past few months. (See Chart 10.) That also points to increased suburban versus city demand.
Chart 10: US Single-Family Home Share (%)
That change in demand is already impacting home prices. The national Case-Shiller house price index has shown relatively stable annual growth in recent months, at around 4.3%. But growth on the 10-City index dropped back from 3.3% in March to 2.8% in June. (See Chart 11.)
Chart 11: Case-Shiller House Prices (% y/y)
But we doubt that growth differential will persist for number of reasons. First, as set out above, we expect demand for city homes in the US to stabilise once the initial wave of those bringing forward a planned move subsides. Second, the change in relative prices triggered by shifting demand will itself act to offset any fall in demand for city living.
Finally, changes in relative prices will trigger a supply response. Land is more plentiful outside of cities, and increased construction of single-family homes will eventually help take some pressure off prices. Indeed, the jump in single-family demand in the US has already helped push homebuilder confidence to record high, and building permits are close to 13-year highs. (See Chart 12.)
Chart 12: Single-Family Building Permits & Homebuildier Confidence (%)
Sources: Census Bureau, NAHB
Increased demand for larger, higher quality homes
Alongside a change in location, the trend toward working from home may change the type of home being demanded. After all, spending significantly more time at home may change housing needs. Most obviously, demand for extra space may rise. But beyond that, demand for higher quality homes and apartments with more amenities, such as gardens or gyms could increase. It is even possible that new homes, perceived to be cleaner, will become more desirable compared to existing homes.
But if a move toward building larger homes were to occur, that would represent a reversal of recent trends in the US. Both the overall size, and number of bedrooms in new single-family homes, has been falling back. From 47% in 2015, the share of new single-family homes completed with four or more bedrooms had declined to 43% by 2019. And the median new home size dropped from 2,467 sq. ft. to 2,301 sq. ft. over the same period. (See Chart 13.)
Chart 13: Single-Family Home Size & Number Bedrooms
Source: Census Bureau
But, barring a small dip during the financial crisis, the overall trend over the past 40-years in both the US and UK has been for larger homes. We suspect that homes have been getting smaller in recent years because land and labour shortages have pushed up construction costs, and lots have been increasingly hard to find. (See Update.) With buyers’ budgets constrained by relatively tight mortgage lending standards, developers therefore cut back on floor area in order to supply profitable homes that buyers could still afford.
Constraints will limit larger single-family homes
For new homes to get larger, those constraints will need to be eased, or house prices will have to rise. It is possible that construction costs will ease over the next decade. Indeed, with unemployment set to remain elevated, the coronavirus may help ease labour shortages in the medium term.
There are also reasons to be positive about the lack of lots. Land is harder to come by in the cities compared to the suburbs, so a shift to building away from cities may help ease land shortages.
That said, lots will still be hard to come by in prime suburban locations near transport links, where local opposition to new development is likely to be significant. Moreover, interest rates are now set to stay lower for longer. Like all assets, other things equal that will raise the price of land.
Higher house prices are certainly a possibility, as wealthy households move to the suburbs and people are willing to dedicate more of their income to housing. But, as we set out further below, there is a limit to how high prices can go.
But multifamily units will get larger
In the US multifamily sector, apartments have also been getting smaller, and that trend started earlier. From 73% in 2006, the share of apartments with two or more bedrooms had dropped to just 50% by 2019. (See Chart 14.)
Chart 14: Multifamily Units with 2+ Bedrooms
Source: Census Bureau
For those who still wish to live in cities for benefits other than proximity to work, there is likely to be increased demand for larger units with extra bedrooms that could function as a home office. That could lead to a reversal in the trend toward smaller units. In addition, there is likely to be increased desire for amenities such more communal outdoor space or balconies. Under our assumption that the coronavirus is no longer a threat beyond the next year or so, there may also be increased demand for amenities such as business centres and shared office space.
Set against that, if the fear of a future infection is still raised, shared spaces will be less desirable. Indeed, if concerns around future infections are permanently higher, demand for amenities to reduce that risk, such as more elevators, enhanced HVAC systems and touchless technology may increase. Once the current infection has passed however, we doubt those concerns will dominate, and willingness to pay more for a slightly less crowded elevator will be low.
In contrast to the single-family sector, we suspect there will be latitude for multifamily units to get larger in the US. In the short term, the drop in city populations will allow for some households to find larger apartments.
Beyond that, while populations are likely to stabilise, there will be a boost to supply from empty office space. As set out in our Global Property Focus, demand for US office space is set to drop back markedly thanks to increased working from home. Accordingly, we think office vacancy rates could rise to 25% by 2025. (See Focus.) It will take time to convert empty offices to apartments, and many will not be suitable. But we expect the share of new apartments in the US with two or more bedrooms will return to between 65% and 70% over the next five to ten years.
There will also be pressure for larger rental units in the UK. However, given the small size of the institutional build-to-rent sector, we suspect the supply response will be more muted.
Income will act as a constraint on prices and rent
With households spending more time in homes and apartments, they may be willing to spend more when purchasing or renting. In addition to the relative change in house prices between urban and suburban locations discussed above, that could act to raise overall house prices and rents. However, any increase in demand will be constrained by affordability, and there are several reasons why the increase in demand probably won’t trigger a fresh boom in house prices.
For one, housing costs are already a large share of incomes. Indeed, one of the features of global housing markets is that prices tend to rise in line, or faster than incomes. This suggests that demand for housing is already high, and that the prices that buyers are paying are already at the limit of what they can afford.
Credit availability will also constrain further growth in prices. In the US, regulations brought in after the financial crisis put a limit on how much income buyers can commit to their mortgage payment. Lenders are also more cautious and now demand higher credit scores. Credit availability will also constrain further growth in prices. In the UK, falling interest rates have already pushed house prices to a high level, and there is limited room for buyers to take out larger mortgages to bid up prices further.
The story is similar in the rental sector. A prolonged period of weak income and strong rental growth in the US since the financial crisis had pushed rents as a share of income to a record high by early 2018. (See Chart 15.) It has since flattened out, and we doubt there is much latitude for households to devote more of their income to rents.
Chart 15: US Rent as Share of Avg. Earnings (Index, 90-15 Avg. = 100)
Sources: Refinitiv, Capital Economics
Admittedly, the reduction in travel costs from increased working from home suggests that homeowners and renters may have a little more disposable income to spend on housing. This could support both rents and house prices, although in most cases we would not expect to see a significant impact on housing affordability.
Beyond that, firms may end up passing on some of the savings from office space to workers’ wages. But that is far from certain. It is equally, if not more likely, that firms will instead choose to distribute any gains from lower office costs to their shareholders. We are therefore sceptical that the shift to working from home will have a large upward impact on house prices. Particularly given the anticipated supply response outside of cities in response to higher demand.
That said, the current surge of households leaving cities due to cohort effects, coupled with low inventory and government income support, have been boosting house prices at a time when unemployment is also high. Outright falls in house prices will therefore be avoided over the next couple of years. Looking further ahead, record low mortgage rates and supportive government policy mean house price growth is set to at least match earnings growth in the UK and US over the next few years. (See Chart 16.)
Chart 16: US & UK House Price Forecast (% y/y)
Sources: Case-Shiller, Nationwide, Capital Economics
We also expect households to commit a similar share of their earnings to rent. In the UK, combined with near-term weakness driven by the direct economic impact of the virus, that implies rental growth will be weaker than our previous forecast.
In the US, given with the move toward larger apartments, that implies rent per square foot will eventually fall back. Coupled with our view that rental yields will fall toward 4.1% thanks to prolonged low interest rates, that means apartment capital values will see a marginal decline. A fall in values is consistent with the rise in supply we are forecasting due to office conversions. Overall, after a slight recovery in values following the direct impact of the virus, we expect capital value growth will average -0.4% y/y from 2025-2029. (See Chart 17.)
Chart 17: US Apartment Capital Value Growth (% y/y)
Sources: MSCI, Capital Economics
In summary, we expect the shift to working from home caused by the coronavirus will impact the residential market in several ways. In the short term, there will be some exodus from cities. However, that will primarily reflect demographics, as a bulge of people currently in their late-20s bring forward a planned move to the suburbs.
But, beyond that, a requirement to go into the offices at least a few days a week will prevent a large drop in the urban productivity or amenity premiums, and cities therefore won’t be deserted.
With households spending more time at home, there may be some increase in the share of income devoted to mortgage or rental payments. However, affordability will limit the impact on house prices, and beyond the short term we doubt there will be a significant change in growth in the UK or US. (See Table 1.)
But there will more of an impact in the rental sector. Many of those living in cities will need more space, and apartments have been getting smaller for years. The good news is that in the US there should be plenty of empty office space which can be converted to apartments. But households will not be able to pay more for bigger apartments, and in the US that implies a decline in both rents and apartment capital values relative to the pre-COVID situation.
Table 1: Pre and Post-Coronavirus Housing Market Forecasts (% y/y, Period Average)
Source: Capital Economics