Yes, Renting Is Obviously More Expensive Than Buying (in the UK)

Ben Gardiner
Sep 5, 2018 · 11 min read

Summary

  • Economic commentators and the ONS frequently claim that renting and buying are equivalent economic decisions – typically suggesting that consumers can freely choose between renting or buying, then showing that annual rental yield isn’t much more expensive than mortgage costs plus depreciation
  • However, UK consumer preferences, investment behaviour and survey data all show an overwhelming preference for owner-occupation rather than renting – in standard economic theory, this would indicate that renting is an inferior state to owner occupaation
  • Since the consumer behaviour is for renters to try to become owner occupiers at all costs, we can discount that rental yields vs asset prices are balanced by the decision to rent vs owner-occupy – they’re balanced by the decision to be a landlord vs an occupier
  • Additionally, since housing is a real asset, and the rental yield rises with inflation, the owners of housing assets receive an inflation-related return each year which renters do not – this is missing from the standard accounting, and means deposits receive an implausibly high gross yield
  • Finally, UK tax law exempts owner-occupied housing from tax on imputed rent (the return an owner gets from owning their house) and any capital gains. As a result, if you consider imputed rent to be part of gross income, the tax burden on owner occupiers is significantly lower than that on renters
  • After these adjustments, renting is significantly more expensive than owner-occupying a property, more than can be justified by a normal risk-adjusted return to the capital invested by the owner occupier
  • Policy makers and economists need to bear in mind this significant difference in cost of owner-occupation and private rental, as it will be a crucial driver of the success or failure of any policy targeted at private renters and the asset-poor

1. Why do some people think that renting and owning are equivalent economic decisions?

If you own a house, you have rights to the benefits of owning the house – both any capital gain and the benefit of being able to live in the house. If you didn’t own the house, to get the same benefit you would need to rent somewhere else of equivalent quality; alternatively, you could choose to rent the house and receive the rent of the house you live in as income.

Indeed the assumption that the cost of owning equals the cost of renting — or rental equivalence — is a core assumption behind the ONS’s new CPIH measure of inflation. To quote the Deputy National Statistician:

“[Rental equivalence] values the ‘housing services’ you get from your house by looking at what it would cost to rent an equivalent property. This is based on the idea that if you did not own a house, you would need to rent one. Or alternatively, by living in the house I own, I am forgoing the rent I could charge. Therefore the equivalent rent is a good proxy for the costs of housing for owner occupiers.”

Clearly there are costs which come with owning a home: the buildings decline in quality, and the costs of this depreciation are borne by owners but not by renters; you will also have to pay mortgage interest, and will not receive the income that you would if you invested your capital in alternative assets. These are costs not borne by renters, who don’t have to pay upkeep and don’t have to invest any money in the building.

Ian Mulheirn of Oxford Economics describes this model in detail and puts empirical flesh on the theoretical bones of this model at the following link:

He uses estimates as follows:

  • rental yield of 4.4%
  • annual costs of a 90% LTV mortgage at 2.4% (which should be divided into a risked and unrisked portion – currently 0.8% from the two year rate and 1.6% from a risked return)
  • depreciation costs at about 0.8%
  • taxation costs from stamp duty at 0.13%

This leaves a residual amount which he terms “forgone capital income”, the return to the deposit used for the house.

Under this estimate, the annual return from rental income on the equity is about 10%, a fairly low return on an asset with annual variance of 40% (4% unlevered).

2. Does the UK consumer agree with this analysis?

Housing ownership is fairly prevalent in the UK, with 64% of the population owning their own home (2015). However, this doesn’t tell the full story. The proportion of owner occupiers among older age groups, which are more likely to be wealthier, approaches 80%, and UK households are unusually likely versus international comparators to hold wealth in the form of property. In fact, UK households show a big preference for holding liquid wealth in the form of housing:

Survey data back this up. In the BRMB/MORI research on preferred tenure cited by the Joseph Rowntree Foundation ( https://www.jrf.org.uk/sites/default/files/jrf/migrated/files/housing-markets-volatility-full.pdf ), 94% of owner occupiers wish to continue to be owner occupiers in 2 and 10 years time, while among private renters 54% and 79% aspire to become owner occupiers over the same time frame. Social rental tends to be significantly cheaper, less prone to price hikes and more secure than private rental, but even among social renters these figures are 31% and 51%.

Savills, the estate agent, also commissioned research on private tenants, asking them why they chose to rent rather than buy. When given the option, only 24% of private tenants chose “flexibility” as one of the reasons they choose to rent, with a large majority of private tenants citing various affordability measures as reasons they have to rent:

http://www.savills.co.uk/research_articles/229130/172142-0

3. So how should we think about renting vs buying?

The above indicates that privately renting is not their preferred choice, so it’s clearly not an equivalent economic decision to buying for the occupiers of the property.

If there is a balance between the economics of renting vs buying for the owners of the property, it is between the different owner types – there should be an equivalence between after-tax and after-costs decision for a landlord vs owner-occupier to buy a property when they are in competition. Focusing analysis on how these owner economics drives occupier economics in for properties with different tenures is likely to be much more sensible.

Using this lens may also provide a better theoretical justification for why different states and countries have very different tenure mixes, as the tax treatment of owner-occupation, private renting and being a landlord can vary significantly.

4. Are we missing any other returns might owners receive?

Rental inflation is crucial, and is missing from the standard analysis. The cost of new rental leases have tended to increase through time. In many areas, housing supply isn’t particularly flexible, and as the price people are willing to pay for housing changes (either due to changes in demand or nominal wage increases) costs should change. For this reason, rents tend to rise at least in line with price inflation, and more often historically in line with wage inflation. If rents rise from one period to the next, a renter will have to pay an increased rent in the following period; an owner occupier will both pay and receive a higher imputed rent. As a result, if we view the price of the house as the discounted value of the future rents, the owner of the asset accrues the value of the rise in rents during each period.

This rental-inflation return isn’t included in the standard analysis, and it makes a huge difference. Some UK Government bonds are built with inflation protection, as they pay realised inflation plus a yield, and by comparing the yields of these bonds to bonds without these protections we can see what inflation is expected to be in the future. Bank of England data has a market implied UK annual inflation rate of 3% over 2 years and up to 4% over 10 years.

Using the lower of these figures in conjunction previous figures, the annual yield before costs of owner-occupied housing is in expectation 7.4%, with forgone capital income of 4%. This means the deposit has an levered return of 40%, two times the return of similarly risky positions in equity or corporate bonds – equity has a risk premium of about 5% and an annualised risk of about 10%, so a similarly risky position with 40% risk will have a return of 20%.

Note: Returns for landlords are lower, because landlords are taxed on both rental income and capital gains, and unlike other investments property investments cannot be put in ISA wrappers. Additionally, they may have void periods and have higher mortgage costs than owner occupiers. This reduces the return of these assets, so the return of households as rental assets is likely similar to that of other assets.

Why, if occupying is more attractive than renting out, do landlords still own and rent out? I suspect different types of occupants will pay different rents, making the landlords the marginal buyer. For example, three lone adults may be willing to pay more rent than a family of four for a house in certain locations: such houses will likely be bought by prospective landlords and families are priced out, even after their improved tax treatment.

The lack of any consideration of inflation in the pricing equation of housing is likely due to the fact the largest and best researched housing market in the world, the United States, often has rent controls, so this is not relevant for consideration there.

Academic backing for the inclusion of an inflation return is in the following NBER paper, Owner-Occupied Housing As A Hedge Against Rent Risk (http://www.nber.org/papers/w9462) – the inflation return we are including is this hedge paying out each period, and I am arguing above that this hedge is too rich.

5. But aren’t these returns coming in tax-free, too?

Yes. The UK tax code applies no taxation to imputed rent. It also excludes primary residences from capital gains tax, meaning that the inflation-linked portion of the gain is not taxed. By contrast, renters are pay their rent from taxed income, received at their marginal tax rate.

For example, take the average first-time buyer in the following link:

With an average household income of £36,000, they are buying a home worth £180,000 with a deposit of £20,000. That home has annual costs of £6,100, an average imputed rent of £8,000 and an inflation-linked return of £5,000.

The equivalent renter has shelter costs of £8,000, no inflation linked return and would have to invest their deposit in public markets, where equivalently risky investment to the deposit would earn about 20%, or about £3.6k (untaxed if invested in an ISA). The overall difference in earnings for the year is £3.7k before tax, or a 14% increase in after-tax income.

In addition, this difference is untaxed. If they were a single earner, as 70%, are, their marginal tax rate would be 32% including national insurance, so to earn similar expected after-tax net pay, their gross salary would have to be £5.4k higher at £41.4k.

The magnitude of this difference becomes very clear when you compare the effective taxation rates for the owner-occupier and the renter with and without an investment:

Investing the capital in property is as valuable an increase as having it at all. Moderate differences in taxes paid driven by differential capital vs labour taxation then become big differences in taxation once magnified by the difference in shelter costs, making it even harder for those not on the property ladder to accrue the capital necessary to get onto it.

Note: taxation regimes in other countries differ hugely. In the US, for example, capital gains on primary residences is taxed heavily; there are also very significant property taxes which are incident on the owners of property (unlike in the UK where smaller taxes are incident on the occupiers of property). In regions where these property taxes rise with inflation but rent control is weak, such as New York, private tenure is generally preferred by both owners and occupiers; in regions where these property taxes are fixed for owner-occupiers and rent controls are strong, owner-occupation is preferred and fewer properties are let out.

6. So what does this all mean?

Renting, at least in the UK, is clearly more expensive than owning property. Saving (or, more frequently, being gifted) a deposit large enough to buy a house with suddenly allows you to earn twice the return, given its riskiness essentially doubling the net present value of the saved amount.

As a result, people in the UK hugely bias towards property investment over any other kinds of investment. Given high marginal taxes on labour vs zero taxes on capital and the relatively shallow middle of the income distribution, it may actually be preferable for most to inherit property rather than human capital, potentially leading to a society of slugs and snails – those born with houses and those without.

If all this is true, no improvements to tenure, contract uniformity or other conditions of private renting will significantly change the tenure preference of people in the UK. Without changes to the taxation of residential property, returns to being an owner-occupier are so high that private renters will always want to become owners. Deposits would remain extremely important in allowing people to get on the housing ladder. In my personal opinion it’s really very important for policy-makers and those involved in the housing supply and tenure debate to understand this and incorporate it into their thinking.


Addendum: Where does this leave ONS data?

ONS data is designed to answer a specific question: how the annual cost of renting has changed year-on-year. The reason for this is because most inflation and consumer-targeted macroeconomic statistics are designed to assess changes in the revenues and costs of households. In finance terms, this is analagous to providing data about households’ earning statements.

In the case of the housing component of CPIH, if the hedonic adjustments designed to ensure that the measure reflects a constant quality of housing are being made correctly, measuring changes in private rents should provide a good answer to this question. In the case of the UK, there have been significant regional divergences in wages, driving significant regional changes both in the quantity and price of rental property supplied. This is adjusted away in the CPIH rental methodology, and this modelling choice (intended to ensure hedonic constancy) in my opinion likely makes it impossible to compare the IPHPR to simple national earnings averages. This will have to be expanded in a further piece.

What ONS data and ONS models are not designed to do are to incorporate changes in the balance sheet of consumers. This is why the model ignores expected return from rental inflation when looking at costs for owner occupiers. Unfortunately, in a country where most households are leveraged asset-owners rather than renters, a wages-driven shock in rents would drive improves most households balance sheets as the discounted housing asset would rise versus the unchanged liabilities. As a result, modelling prosperity, rather than just costs, requires better and more detailed models.

These issues, however, are not necessarily issues with the ONS calculation methodology: they are more important caveats for economists and policymakers to bear in mind when using this data series for purposes different to its original design.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade