The negative economic impacts of COVID-19 have bypassed the housing market so far due to the extraordinary support measures put in place by the government, regulators, and banks. But while policy has probably reduced and delayed the impact of COVID-19 on house prices, it has not removed it altogether and the pandemic will take its toll in 2021. That said, the correction will be far smaller than in previous recessions. Indeed, the 5% dip in prices we now expect next year would leave prices marginally higher at the end of 2021 than they started 2020. And if there is a swift recovery in jobs as we anticipate, there is every prospect of a return to growth in 2022.
- The negative economic impacts of COVID-19 have bypassed the housing market so far due to the extraordinary support measures put in place by the government, regulators, and banks. But while policy has probably reduced and delayed the impact of COVID-19 on house prices, it has not removed it altogether and the pandemic will take its toll in 2021. That said, the correction will be far smaller than in previous recessions. Indeed, the 5% dip in prices we now expect next year would leave prices marginally higher at the end of 2021 than they started 2020. And if there is a swift recovery in jobs as we anticipate, there is every prospect of a return to growth in 2022.
- House prices have confounded forecasters expectations this year by shrugging off the economic hit from COVID-19 and rising by about 6% y/y, the largest increase since 2014.
- That’s because the usual channels through which a recession hits the housing market, of rising unemployment and mortgage payment difficulties, have been mitigated by the furlough scheme, mortgage payment holidays, and a moratorium on repossessions.
- Meanwhile, the market has been boosted by pent up demand from the first lockdown, a revaluation of space needs due to working from home, and an extra kick from the stamp duty holiday.
- But the policy support that has protected and boosted the market this year is due to be withdrawn in 2021, just when we expect the unemployment rate to peak at 7%. The housing market has never escaped unscathed from a drop in employment of the scale we forecast. In fact, in isolation the historical relationship between employment and house prices suggests a 25% house price crash is in the offing. Our view, however, is that a fall closer to 5% y/y in Q4 2021 is more likely.
- There are three strong reasons why we don’t think 2021 will as bad as 2008-09 or the early 1990s.
- First, the drop in employment will continue to be concentrated on those that rent rather than own their homes.
- Second, while housing is arguably overvalued, due to very low interest rates, mortgage payments are very affordable and will remain so.
- Third, banks are in a strong financial position and will continue to provide generous forbearance and not dramatically cut the supply of mortgage credit.
- Taken together, these factors suggest we will see a much more modest rise in forced sellers than in 2008 or 1990, so a severe correction should be avoided.
- There are both upside and downside risks to our forecast. If government support is more generous than we assume and the stamp duty holiday and furlough scheme are extended, a fall in house prices might be prevented altogether. Alternately, if banks and households start to expect a large fall in prices, it can become a self-fulfilling prophecy.
- Ultimately, we expect the economy and employment to recover swiftly. And with interest rates set to remain very low, that will prevent a prolonged fall in prices. So, after a dip in prices in 2021, we expect them to start rising again in 2022.
House prices set to dip in 2021
The negative economic impacts of COVID-19 have bypassed the housing market so far due to the extraordinary support measures taken by the government, regulators, and banks. While policy has probably reduced and delayed the impact of COVID-19 on house prices, the pandemic will take some toll on them in 2021. But the correction will be far smaller than in previous recessions for several reasons. First, the drop in employment has been focused on renters rather than homeowners. Second, housing wasn’t wildly overvalued when the crisis hit. Third, mortgage payments are affordable. And finally, banks are in a good position to continue supplying mortgage credit.
2020 is expected to be a very good year for prices
Many people will be glad to see the back of 2020. The housing market is an exception. It has not only proven resilient to the economic hit from COVID-19, but rallied much sooner than anyone expected. Annual house price growth hit 5.8% in October, suggesting it will average about 6% in Q4 which would mark the largest rise in prices since 2014. (See Chart 1.)
Chart 1: House Prices (Q4 %y/y)
Sources: Refinitiv, Nationwide, Capital Economics
That surge in prices is in stark contrast to what forecasters, including us, expected in the spring. When it became clear that COVID-19 would cause much of the UK economy to be shut down, the consensus for house price growth this year fell to minus 3%. (See Chart 2.) Since then, of course, the government and the Financial Conduct Authority (FCA) have stepped in with multiple policies to mitigate the impact of the sharp reduction in GDP on the wider economy and the housing market. At the same time, pent up demand from the first lockdown, a revaluation of space needs due to working from home, and the extra kick from the stamp duty holiday have led to a surge in activity and prices.
Chart 2: Evolution of House Price Forecasts for 2020 (Q4 %y/y)
Sources: HMT, Capital Economics
But COVID-19 is yet to take its full toll
Despite the surge in public debt, the policy response can surely be deemed a success. And with the government seemingly on course to vaccinate most of the population by Q2 next year, we now think that GDP will recover to its pre-virus level by early 2022 rather than in 2023 as we previously expected. However, some businesses have already gone bust, hundreds of thousands of people have already lost their job, and the economy will need to work through some structural shifts.
What’s more, 3.3 million were still on furlough at the end of August and this total probably increased when the economy went into the second lockdown in November. Sadly, we suspect a substantial proportion of furloughed workers will not return to their old jobs even if restrictions are gradually eased next year as we assume. The cumulative drop in employment is likely to grow from 566,000 in September to 1.7 million in Q2 2021. And while we no longer think that the unemployment rate will rise as far as 9%, we still think that it will increase from 4.8% now to 7%. (See here.)
As a result, while the hit to the housing market from the virus has probably been reduced and delayed, we doubt it will be avoided completely. Policy support will not continue indefinitely. The moratorium on involuntary repossessions, that has caused them to drop close to zero despite the rise in unemployment, is due to end in January 2021. (See Chart 3.) The furlough scheme and the stamp duty cut will both end in March 2021. And mortgage payment holidays, of up to six months, will end by July at the very latest.
Chart 3: Unemployment & Possesions
Sources: Refinitiv, UK Finance, Capital Economics
Meanwhile, the stamp duty holiday has brought buyer demand forward from 2021 into 2020 which creates problems. We have already noted that with hindsight, the cut was unnecessary. (See here.) And it now looks like the boost to demand will be most needed in 2021 after other policy support is withdrawn. At the end of the much less generous stamp duty holiday in 2008-9 there was a clear change in momentum in house prices. (See Chart 4.) We expect to see the same phenomenon this time around.
Chart 4: House Prices When Stamp Duty Cuts Expire (% 3m/3m)
Sources: Nationwide, Refinitiv, Capital Economics
When all that support is withdrawn, the 4% peak-to-trough drop in employment we forecast will take its full toll on the housing market. House prices have never escaped unscathed from a fall in employment of that scale. (See Chart 5.) In fact, looking at the relationship between employment and house prices in isolation suggests a drop in prices of over 25% could be in the offing.
Chart 5: Employment & House Prices (% y/y)
Sources: Refinitiv, Capital Economics
Three reasons why house price falls will be limited
While we suspect that house prices will reverse much of their 2020 surge next year, we are not forecasting a correction in house prices implied by the historical relationship between house prices and employment for three main reasons.
The composition of unemployment
The drop in employment so far has impacted more heavily on renters rather than owner occupiers. (See here.) Because it has led to the closure of shops, hospitality venues, and the cancellation of events, COVID-19 has hit the retail, leisure and hospitality industries hardest, which employ a large number of younger and lower skilled workers who tend to rent. Indeed, the fall in employment so far is accounted for entirely by a drop in employment in “routine” and low skilled work. (See Chart 6.)
Chart 6: Change in Employment by Occupation (Q1-Q3 2020, 000s)
Sources: ONS, Capital Economics
Meanwhile, employment among highly skilled workers, who are more likely to own their homes, has continued to rise this year. As a result, the drop in employment so far has had less impact on owner occupiers.
We do expect this to change when the furlough scheme ends. It makes sense that employers have kept highly skilled employees on furlough for longer as they are harder to replace. And we know most of those on furlough are aged between 35 and 64, an age at which many will have mortgages, whereas the drop in employment so far has been focused on under-24-year-olds. (See here.) Nonetheless, the composition of the rise in unemployment is likely to remain tilted towards renters. That means it should lead to a smaller drop in buyer demand and fewer mortgages going into arrears than we would usually expect given the size of the increase in unemployment.
Valuations & Affordability
House prices are arguably somewhat overvalued, but not to the same extent as in 2007. The house price to earnings ratio is close to its all-time high. While much of this is concentrated in London and the South East, comparing the house price to earnings ratio to rental yields suggests houses are somewhat too expensive as well. (See Chart 7.)
Chart 7: House Price to Earnings Ratio & Rental Yields
Sources: Refinitiv, Capital Economics
But, in stark contrast to 2007, because of very low mortgage interest rates, mortgage affordability is very good. And our forecast is that interest rates will stay at 0.1% for the next five years, so affordability shouldn’t deteriorate anytime soon. At current mortgage interest rates, a £180,000 mortgage with a 25-year term would only take 36% of the median full-time wage to service. Ahead of the 1990 and 2008 house price corrections, the typical mortgage would have taken up over 60% of disposable income each month. (See Chart 8.)
Chart 8: Typical Mortgage Servicing Costs (% Median Disposable Income)
Sources: Refinitiv, Capital Economics
Better affordability is another reason to think that we won’t see anywhere near as big a surge in forced sales as in the financial crisis or early 90s. Greater headroom to continue to service the mortgage for households that experience a fall in their income should prevent arrears increasing as quickly. And more forbearance options, like extending the term of the mortgage to lower the monthly repayments, could be sufficient solutions to keep households in their homes.
Banks can help rather than hinder
Banks have already provided generous forbearance by granting 2.5 million mortgage payment holidays and halting involuntary repossessions. And large capital buffers and healthy mortgage books mean they can continue to be generous after the temporary guidance from the FCA expires.
The capital buffers banks built up since the financial crisis are now large enough to withstand an increase in the unemployment rate to over 15% according to the Bank of England’s calculations. So loan losses resulting from the recession so far and the fall in employment that we expect to come shouldn’t lead to banks drastically reducing credit availability. That should mean that fewer arrears are converted into repossessions.
And mortgage books are resilient to house price falls. At the end of 2019, over 70% of outstanding mortgages had a loan-to-value (LTV) ratio below 70%. (See Chart 9) In the early 1990s and 2008 some repossessions occurred because borrowers were unwilling to continue to pay off their mortgage, rather than being unable to, as by doing so they would realise a loss. But under the Bank of England’s coronavirus scenario, in which house prices fall by 16%, only a small share of banks’ mortgage books would rise above 100% LTV. Seeing as house prices have risen 6% since the end of 2019, prices would now need to fall by over 20% for even a small proportion of loans to fall into negative equity. So it is very unlikely to become an issue, which is another factor pointing to a modest increase in possessions.
Chart 9: Effect of Stress Scenarios on UK Banks’ Mortgage Book LTVs (% of Book)
Source: Bank of England
As a result of higher affordability than in the past, well capitalised and more accommodating banks, and little risk of negative equity, we expect the conversion rate between arrears and repossessions to be much lower than in the financial crisis. This means we should only see a modest increase in forced sales. (See Chart 10.) A big rise in distressed sales was an important factor in the last two major house price corrections. The much smaller increase we expect is an important reason why house prices will not fall by as much this time around.
Chart 10: Arrears & Repossessions (% Outstanding Mortgages)
Sources: UK Finance, Capital Economics
While the fundamentals suggest that any fall in house prices will be limited, house prices can become divorced from their underlying drivers when sentiment is overly downbeat or exuberant. If households and banks begin to expect a big fall in prices, it can become self-fulfilling.
Indeed, we have already seen banks raise interest rates on high LTV mortgages. (See Chart 11.) This is partly a response to very high demand. (See here.) But the latest credit conditions survey in October shows it has also been driven by deteriorating expectations for the economic outlook and banks guarding against the potential fall in house prices ahead.
Chart 11: Quoted Mortgage Interest Rates
Source: Bank of England
Ultimately though, we don’t think we will see a prolonged fall in house prices because we expect the economy to recover swiftly once a vaccine is rolled out. (See here.) Indeed, in our latest forecast employment recovers to its pre-virus level by Q3 2022. This means the financial distress for homeowners will be much briefer than in the financial crisis, for instance, again limiting forced sales. If we are right, downbeat buyer sentiment and banks’ risk aversion is unlikely to last long.
And there is a possibility that government support is more generous than we assume, in which case house prices may not fall at all. If the vaccine is successfully rolled out, the furlough scheme was extended to late-2021 preventing much of a further fall in employment, and the stamp duty holiday was extended, house prices could avoid falling in the COVID-19 recession altogether.
The expected fall in employment; the end of mortgage payment holidays and the repossessions ban; and the end of the stamp duty holiday mean that house prices are likely to reverse much of their 6% surge in 2020. In our view, a fall of 5% y/y in Q4 2021 is now likely. That will be an abrupt turnaround after 2020’s mini boom, but a far smaller decline than that scale of drop in employment suggests.
Indeed, coming on the heels of the jump this year, prices would end 2021 marginally higher than they started 2020. (See Chart 12.) And if the economy recovers as swiftly as we expect, against a backdrop of very low interest rates, prices are likely to start rising again in 2022. We think that house price growth will reach 2% y/y by the end of 2022.
Chart 12: Nationwide House Prices
Sources: Refinitiv, Nationwide, Capital Economics
Andrew Wishart, Property Economist, +44 (0)7427 682 411, email@example.com