Mariana Gimpelewicz and Tom Stratton.
Who is living in private rental properties, and why? The buy-to-let market has been headline news recently. Typically the story has been profit-hungry landlords squeezing out first-time buyers. But landlords are only half of the story. This post examines the rental market from the perspective of tenants. Our work suggests demand for private rental properties cannot explain all of the growth pre-crisis, but the case for over-exuberance is inconclusive. We think that factors driving tenant demand, including demographics, social housing and credit availability, accounts for around half of the growth in the Private Rental Sector (PRS) pre-crisis and over 80% post-crisis. The most important driver post-crisis has been tighter credit conditions, which generated demand for an additional 1 million PRS properties. Looking ahead, we project that tenant demand will drive the PRS to swell by up to an additional 1 million properties between 2014 and 2019. If tenant demand were the only factor in play this would translate to annual growth in the number of buy-to-let mortgages of 2-7%.
Background
Since 2008, the number of outstanding buy-to-let mortgages grew by 6.1% per annum on average, while the number of owner-occupier mortgages fell by 1.7%. Rapid growth rates in the buy-to-let market have caught the attention of policymakers. The Financial Policy Committee (FPC) noted the potential implications for financial stability. Meanwhile, the Prudential Regulation Authority recently published proposed guidelines for lenders underwriting buy-to-let loans.
Fast growth in buy-to-let lending is not a threat to financial stability in itself. The FPC has acknowledged that some growth is consistent with an increase in tenant demand for private rental accommodation. This demand will be met by rental properties bought primarily using cash and buy-to-let mortgages, collectively labelled the PRS. So to understand growth in buy-to-let mortgages we need to place it in the context of demand for PRS accommodation overall.
The share of the PRS as a proportion of all properties in the UK has been rising since 2002, financed in part through the expansion of buy-to-let lending (Chart 1). However, little has been done to-date to understand the role of tenant demand in driving this expansion.
Chart 1: Stock of dwellings by tenure and growth
Source: Department for Communities and Local Government, Council of Mortgage Lenders and Bank calculations
Factors that affect tenant demand
We split the drivers of tenant demand into categories:
Demographics matter because there are cohorts of the population, such as students, immigrants and younger people, who are more likely to rent. If these groups expand, all else equal, we expect there to be more demand for PRS properties.
Social housing provision (i.e. owned by local authorities or housing associations) impacts the viability of alternatives to renting. If the provision of social housing falls then people will be funnelled towards the PRS.
Prospective first-time buyers’ ability to secure a mortgage dictates how easily households can leave the PRS. It is determined by two key factors: credit conditions and affordability. Credit conditions are defined by banks’ interest rates and underwriting standards, which include borrowers having to raise a minimum deposit and demonstrating they have sufficient income to service the mortgage. We think of affordability constraints as borrowers’ ability to secure a mortgage given a set of credit conditions. For example, if borrowers cannot meet deposit requirements because house prices are rising faster than incomes, rather than banks tightening credit standards. Our analysis focuses on credit conditions, which we can quantify more readily.
Explaining past growth
Chart 2 shows the number of additional properties we expect in the PRS for a given period, split by the quantifiable drivers of tenant demand (see Technical Annex for more detail).
Chart 2: Quantifying drivers of tenant demand
Sources: ONS, English Housing Survey, Migration Observatory, Council of Mortgage Lenders, Department for Communities and Local Government, Higher Education Statistics Agency and Bank calculations.
- 2002 – 2007: More students and migrants, less social housing
Net inward migration was the largest contributor to demand for PRS properties during the pre-crisis period (2002-2007). We estimate it generated demand for around 290,000 additional PRS properties (blue bars), around a quarter of actual total growth. Higher tenant demand was also fuelled by 150,000 additional students in higher education (yellow bars), and a reduction in social housing that meant 130,000 additional households required PRS accommodation (green bars). Other population dynamics cancelled each other out; the growing population and a smaller average household size were offset by the shrinking number of youngsters, who are more likely to rent (pink bars). Our deconstruction of tenant demand can explain approximately half of the pre-crisis increase in PRS properties (Chart 3), and cannot fully account for the growth in buy-to-let mortgages (red bar).
- 2008 – 2013: Credit crunch!
Migration was also a significant contributor to post-crisis (2008-2013) tenant demand, accounting for approximately 170,000 PRS properties. However, social housing was a much smaller factor, primarily due to the expansion of housing associations (Chart 1). The contribution from students and other demographic trends were also negligible.
Instead, the primary driver of tenant demand post-crisis was tighter credit conditions for first-time buyers. For simplicity, we compare the share of households securing a high LTV mortgage post-crisis relative to pre-crisis, when arguably all prospective borrowers were able to access mortgages. Our estimates suggest that, as a result of the withdrawal of high LTV mortgages, ‘frustrated’ demand for mortgages from prospective home owners meant one million additional households required PRS accommodation, accounting for 66% of the growth in PRS properties during the period (orange bars). Overall we can account for 82% of PRS growth during this period, and all of the increase in buy-to-let mortgages.
Chart 3: How our estimates of tenant demand compare to actual growth in the PRS
Sources: ONS, English Housing Survey, Migration Observatory, Council of Mortgage Lenders, Department for Communities and Local Government, Higher Education Statistics Agency and Bank calculations.
- Unexplained growth
The size of the gap between pre-crisis PRS growth and our estimates using tenant demand factors warrants further thought (grey bar, Chart 3). We believe the unexplained growth is related to housing becoming less affordable. The question for our purposes is whether buy-to-let mortgages generated their own demand by pushing up house prices, pricing out first-time buyers, or if they were meeting tenant demand from those who couldn’t afford a mortgage for other reasons. On one hand, the introduction of buy-to-let mortgages in the mid-1990s and loosening buy-to-let underwriting standards in the early 2000s boosted credit supply, facilitating additional investor demand for housing. However, there were other market forces also putting pressure on prices at the time. Home-movers were capitalising on the rising value of their own properties by withdrawing equity to upscale, while there was a shortage of housing overall. In the context of these other factors, buy-to-let mortgages could have helped fund the expansion of the PRS to meet rising tenant demand. Isolating the drivers described above is difficult and requires further work. Without concrete evidence on the impact of buy-to-let mortgages on house prices, it is not possible to say whether the gap represents a buy-to-let ‘bubble’. Consequently, we concentrate on explaining levels of activity in the PRS and buy-to-let mortgage market.
Projecting PRS and buy-to-let mortgage growth
It is still useful to estimate the growth rates we might expect in future based on tenant demand dynamics alone. Our projection suggests that additional tenant demand for PRS properties could be less than half between 2014 and 2019 compared to the post-crisis period (Chart 4). This translates to an average annual growth for the number of outstanding buy-to-let loans in the range of 2-7%. By comparison, average annual growth for 2014 and 2015 was 7.6%.
The extent to which demand for PRS properties falls primarily rests on whether ‘frustrated’ prospective first-time buyers from the post-crisis period eventually secure a mortgage. We have calculated an upper-bound estimate for demand, assuming no ‘frustrated’ buyers secure a loan; and a lower-bound estimate, assuming all ‘frustrated’ buyers get a mortgage following a 10-year delay (see Technical Annex).
To think about what this means for buy-to-let mortgages we consider the extent to which they have supported PRS growth in the past. Buy-to-let mortgages funded more than 70% of the expansion in the PRS pre-crisis, but only around a third post-crisis. This difference likely reflects tighter credit conditions for buy-to-let borrowers and stronger demand from cash investors following the crisis. Using these historical shares Table 1 sets out the range of growth rates for the buy-to-let market consistent with our projections for growth in the PRS.
Chart 4: demand for the PRS going forward
Sources: ONS, English Housing Survey, Migration Observatory, Council of Mortgage Lenders, Department for Communities and Local Government, Higher Education Statistics Agency and Bank calculations.
Conclusion: have reports of a buy-to-let bubble been greatly exaggerated?
Using drivers of tenant demand we can explain three quarters of the expansion in the buy-to-let market pre-crisis. After the crisis, however, we find that rising tenant demand comfortably accounts for the increase in the stock of buy-to-let mortgages. Indeed, demand resulting from tighter credit conditions for first-time buyers was twice the growth in buy-to-let mortgages during the period.
Looking ahead, there could be up to one million additional properties in the PRS between 2014 and 2019 due to growing tenant demand. Our projections suggest we should expect growth rates for buy-to-let mortgages of between 2% and 7% over this period. By comparison, average annual growth since 2014 has been 7.6%. But this does not automatically confirm claims of a bubble, particularly as we cannot capture changes in demand related to the affordability of housing.
Mariana Gimpelewicz and Tom Stratton work in the Bank’s Macro-financial Risks Division
If you want to get in touch, please email us at bankunderground@bankofengland.co.uk. You are also welcome to leave a comment below. Comments are moderated and will not appear until they have been approved.
Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.
Tenant demand??????? Tenant demand??????????!!!
Buy To Let investors are the housing crisis.
The main reason people can’t afford homes, is the extra demand from 5 million landlords who occupy something like 1 in 5 houses.
Demand for rental accommodation is created by landlords when they outbid owner occupiers, because every time a landlord buys a house they create a tenant.
You can possibly think that a shortage of housing is what forces people to rent houses, can you!?!
Every time a landlord buys a house, they outbid a prospective owner occupier. Every time they outbid a prospective owner occupier they create a prospective tenant.
Tenant demand??!! For the most part, landlords are creating the demand they pretend to service. They provide nothing to the tenant that the tenant would not have had without them.
Yes, there probably is a residual, natural demand, but it would almost certainly be tiny. Holiday lets and student accommodation would probably still exist in a free-market.
Most people who rent do so because they can’t afford to buy, not because they don’t want to, and they can’t afford to buy because landlords are hogging all the houses, and have outbid renter-savers along the way.
Thankfully we now have a stamp-duty hike on the BTL investors, and Section 24 set to be phased in from 2017 – although it should be impinging on market now, for the BTL minds who have made themselves informed about it.
“It’s interesting because it is kind of chapter and verse the BTL investors argument, i.e. the argument by which they typically write themselves out of the mismatch between what first-time buyers can afford and what first-time buyers have to pay if they want to actually buy:
Houses have become more expensive
At the same time credit conditions have tightened
As a result there are lots of people who want to be first-time buyer, but they can’t afford to buy
Fortunately, using the frustrated first-time buyers wages to service an interest-only mortgage a buy-to-let investor can afford to buy
Hence the buy-to-let investor is the good guy, ‘providing’ a house for rent and answering the rental demand created by the high prices and tight credit.
The blog post posits three potential buyers; investors using mortgages (i.e. buy-to-let), investors using cash and first-time buyers.
However, surely if buy-to-let credit underwriting was so tight that the buy-to-let investor couldn’t outbid* the household that wished to become homeowners using the households own wages (the BTL investor and FTB are both really bidding with the the households’ wages expressed via a mortgage) then prices would fall back in line with wages (as expressed through the prevailing credit conditions offered to first time buyers using mortgages, which would be lower). If you took the BTL investors out of the picture, the only way this fall back in prices wouldn’t happen would be if cash-financed investors could take up all the slack at current prices – i.e. continue to buy their current share and buy the share currently being purchased by BTL investors too.
It seems to me that the Bank of England shouldn’t be complicit in pricing people out. Hence regulation of lending needs to remove the situation where BTL investors are capable of pricing out people who want to buy (using the wages of the person who wants to buy). If BTL investors are taken off the board as setters of prices and cash buyers rush in to fill the gap and prices still don’t fall, then solving the issue of housing affordability becomes a matter for the Treasury to be tackled by taxing investments in residential property (and increasing supply).
Essentially, the blog is arguing that it isn’t the availability of BTL mortgages on present terms that is causing the PRS to grow so rapidly. However, the argument strikes me as tendentious, to say the least. As the Bank of England is responsible for implementing the Mortgage Market Review at a time when buy-to-let financing is so different to owner-occupied financing it is deeply problematic if the Bank of England employees are also offering thought pieces where they implicitly suggest that the existence of buy-to-let financing has no impact on prices.
The blog is written as if prices have a level which is largely independent of the terms of credit, i.e. tight credit is accepted as a source of rental demand, because potential FTBers can’t afford to buy, but the role of loose credit in setting those prices is ignored. A crucial aspect of the setting of prices is the existence of buy-to-let lending, which is of course an aspect of the prevailing credit conditions.
At present if you want to move home you do not need to find a willing buyer. You can just take out a buy-to-let mortgage. As a consequence, the rent achieved by the ‘old home’ and prevailing BTL credit conditions imply a price for the house out of which you wish to move. If a buyer willing to pay that price cannot be found, you just take out a buy-to-let mortgage. Hence anybody who wants to obtain that house from you must pay more than the price implied by the rent and the prevailing BTL credit conditions.
To consider a market place where this mechanism is present and argue that the PRS expansion merely answers rental demand from frustrated first time buyers seems to write credit conditions out of a story which could easily be very, very heavily influenced by the credit conditions offered to buy-to-let investors.
* After all if two people want the same thing and only one afford the market price the act of buying is implicitly an act of outbidding.”
There is one other consideration which may be important for households with low and uncertain earnings. Owner-occupation is risky, since an unanticipated downturn in income may lead to repossession (which has high pecuniary and psychological costs). However the housing benefit system acts as an insurance mechanism for households who are in the PRS. It seems very plausible that the general increase in economic insecurity after the 2008 crisis led many renting households on the margin of owner-occupation to abandon or defer plans for house purchase.
This is a thought-provoking piece. I would like to challenge one of the assumptions made: that “rising tenant demand comfortably accounts for the increase in the stock of buy-to-let mortgages”, particularly with the “priced out” group. Instead, there is a more complex chicken and egg situation and that the growth of mortgaged buy-to-let (BTL) itself accounts for much of the tenant demand.
The article notes that many first time buyers are priced out and this is largely as a result of high house prices and also by strict affordability criteria that banks impose following the FCA’s Mortgage Market Review (MMR). The result of the latter is that a potential borrower must usually demonstrate that they can afford the payments on a 7% repayment mortgage. This is clearly a challenge at current prices.
By contrast, a BTL borrower has typically only needed to demonstrate that a 75% LTV interest-only mortgage, the expected rent exceeds the interest payment by 25% at a stressed rate of 5.49% (although some lender’s conditions vary slightly).
This means that a BTL borrower will always be able to outbid a first time buyer for a property, even if they have the same deposit. For example, with a £150,000 mortgage on a £200,000 property, the potential owner-occupier must show that he or she can meet monthly payments of £1,060 on a repayment mortgage at 7%. By contrast, the BTL investor would only need to meet interest-only payments of £686 at 5.49%.
The only thing potentially that might stop the BTL investor is the rent – it must exceed £686 by 25% i.e. £858. As long as this condition is met, the BTL buyer has the flexibility to outbid the first time buyer.
So what is the “real” value of a property? That is difficult, but I can tell you what the minimum price of a property is 2016 – it is the price at which on a 75% LTV mortgage, the rent exceeds an interest-only mortgage by 25% at a stressed rate of 5.49%. For a property with a monthly rental value of £1,000, this would equate to £233,000. There is therefore no need for a seller to sell for less than this price.
BTL investors are demonstrably able to outbid first-time buyers and clearly have been doing so in great numbers – nearly all of the net increase in mortgage lending post-crisis has been to BTL investors, as noted in the Financial Stability Report last year. Given that the UK has a fairly fixed housing supply, every purchase of an investment property for rent creates ensures that there is at least one more individual who must rent for the foreseeable future. In effect, the very existence of BTL has created the demand that the article suggests that investors are meeting.
Changes to interest rate tax relief and the suggestions in the PRA’s consultation on BTL underwriting standard will go some way to levelling the playing field for first time buyers, but the impact will take some time to be felt.