Microsoft Word - The economic rationality of landlords _cover

Kathleen Scanlon and Christine M. E. Whitehead

The economic rationality of landlords

Conference paper

Original citation:

Scanlon, Kathleen and Whitehead, Christine M. E. (2006) The economic rationality of landlords.

In: Housing Economics Working Group, 2006, Copenhagen, Denmark.

This version available at:

http://eprints.lse.ac.uk/29951/

Available in LSE Research Online: December 2010

LSE has developed LSE Research Online so that users may access research output of the

authors and/or other copyright owners. Users may download and/or print one copy of any article(s)

in LSE Research Online to facilitate their private study or for non-commercial research. You may

not engage in further distribution of the material or use it for any profit-making activities or any

commercial gain. You may freely distribute the URL (http://eprints.lse.ac.uk) of the LSE Research

Online website.

CORE Metadata, citation and similar papers at core.ac.uk

Provided by LSE Research Online

The economic rationality of landlords

February 2006

Paper prepared for the 16-17 February 2006 meeting of the

Housing Economics Workshop, European Network of Housing Research

Copenhagen Business School, Denmark

Authors:

Kathleen Scanlon, London School of Economics and guest researcher, Danish

Building Research Institute

Christine Whitehead, London School of Economics

1 Introduction

UK government policy towards the private rental sector reflects the assumption that small

landlords are less commercially oriented than larger ones. Is this so? This paper uses

data from a large survey of UK investor landlords to ask to what extent the behaviour of

such landlords is economically rational, and whether professional landlords behave

differently from non-professionals.

The principles of investment theory apply to rental property, as to any other investment,

but certain particularities make it difficult to compare returns from real property with

those from alternative investments. Various economic stimuli will affect the returns from

rental property--we categorise these stimuli in terms of how they would affect rental

returns, then make predictions about how a rational investor would behave when faced

with these stimuli.

Most empirical research into private landlords emphasizes the diversity of landlords’

motives for entering and continuing in the property market—profit maximization not

necessarily being high among them. Researchers in various countries have attempted to

explain different landlord behaviour patterns by assigning landlords to categories

according to the number of properties they own, their motives for being a landlord, etc.

Drawing on these classifications, we develop our own, which distinguishes ‘professional’

investor landlords from ‘non-professionals.’

After a brief description of the development of the buy-to-let market in the UK, we

discuss the results of our analysis. This uses data from a 2004 survey of 1300 UK buy-

to-let landlords, which asked landlords how they would react to several hypothetical

economic stimuli. To what extent were their predicted responses ‘rational’? Did the

responses of professional and non-professional landlords differ? Finally, we offer some

conclusions and some observations about what they might mean for policy.

2 General principles: investors as return-maximisers

Most economic analysis of the behaviour of investors, including landlords, starts from the

premise that they are rational economic actors—that is, that their decisions about their

investments (in this case, rental property) are governed by the goal of maximizing their

returns. If rental property is viewed as an investment, then economic theory suggests that

the risk-adjusted expected return on property assets should equal that on other types of

asset, for otherwise investors would shift their holdings—if investors expect the risk-

adjusted return on property to exceed that available, say, in shares, then they would sell

shares and buy property; and conversely they would sell property and buy shares if the

expected return on property were below that available in the stock market (we are

ignoring diversification, for the sake of simplicity).

Recently, however, some behavioural economists have argued that investors do not in

fact rationally assess relative rates of return. Robert Shiller observes that

Economists usually like to model people as calculating optimally their investment

decisions based on expectations of future prices and estimates of the risk in

alternative investments. However, in fact, the typical investor’s actual decision

about how much to allocate to the stock market overall, and into other asset

classes such as bonds, real estate, or other investments, tends not to be based on

careful calculations. Investors are not often assembling forecasts for these

different asset class returns and weighing these against measured risks. (Shiller,

2000, p. 55)

Recent research suggests that property investors in the USA and Australia have not been

acting rationally with respect to potential growth in both rents and stock market returns,

while property investors in the UK appear to behave more rationally over the cycle.

(Hendershott and MacGregor, 2006)

Harvey puts the rational-investor position, in its simplest form:

Anybody who purchases a property rather than renting is an investor. The

satisfaction or return received should at least equal what could be obtained if,

instead, premises were rented and the money invested elsewhere. (Harvey 1981,

p.48)

According to Skifter Andersen

The simple foundation of economic theory…on the investment behaviour of a

landlord is the assumption that he maximizes the current value of the expected

stream of profits (the share of rental income that remains after running costs and

maintenance expenses are paid) arising from the ownership of a rental property.

In view of the extreme durability of housing a landlord is assumed to adopt a

perspective of an infinite horizon. The expected profits in future periods are

discounted in order to obtain their present value. Rational expectations are

assumed. (Skifter Andersen, 1998, p. 181)

The definition of return in the case of rental property is not necessarily simple. Some

researchers, including Skifter Andersen, consider only net income (net annual rental)

return on open market value, while others look at total annual returns (net annual returns

plus capital growth—real or nominal) (Crook, Kemp, Barnes and Ward, 2002). For the

purposes of this paper we are interested in both rental yield (net of costs) and expected

change in capital value. Net rental yield is a known quantity, at least until the end of the

current lease period. Change in capital value is much less certain but probably more

important to most investors, as historically it has, over the long term, outweighed rental

returns. (Many UK landlords who purchased properties in the last few years must be

anticipating that capital gains will make up all of their total return, since their net rental

yields are often negative.) A 1999 study of Australian landlords found that the

expectation of capital gains was as important to them as rental income, if not more so.

(Beer, 1999)

3 Some complications

In a world where all investors behaved as profit-maximisers, landlords might be expected

to increase or decrease their rental portfolios depending on the expected rate of return on

property and that on other investments. However, housing has certain characteristics that

set it apart from many other investments and make asset reallocation more difficult:

Housing assets are neither divisible nor fungible; housing assets are lumpy.

Furthermore, the adjustment and transactions costs associated with housing tend

to be large and lumpy too. In addition, participants in the housing market will

economise on transactions costs by limiting their number of housing transactions

and the liquidity of housing assets will be constrained accordingly. (Baddeley,

2005, p.9)

It can be difficult to compare the returns from property and other investments, for several

reasons:

• The tax treatment of various investment vehicles may differ. There are particular

issues associated with the treatment of property as a perpetual asset and with the

differential treatment of rented and owner-occupied property, which impact on

both investor behavior and the strength of the market.

• Transactions costs must be taken into account. In the UK, transactions costs for

the purchase of residential property are roughly 3-8% of property price on a

purchase, and somewhat less on a sale

. This introduces a significant financial

hurdle to portfolio rebalancing.

Transactions costs on purchase comprise legal fees, estate agents’ fees (usually 1.5-2% of property price),

bank fees associated with securing a mortgage, and Stamp Duty Land Tax. This is levied at a rate of 1% on

• Rental property is often purchased with borrowed money (a mortgage); this is less

often the case with the principal alternative investments, such as shares. Because

the landlord leverages his investment with borrowed money, any increase in

capital value translates to a much higher rate of return on the landlord’s own

investment.

Conversely, a fall in property values can wipe out a highly leveraged

landlord’s equity completely. Also, many landlords depend on rental income to

pay the mortgage, so cash flow is critical. This is not generally an issue with

other investments.

• The probability of potential changes in taxation, subsidy (e.g. housing benefit),

regulation and the economic environment all have impacts on expected returns

but that impact is difficult to quantify and therefore evidence of differences in

returns across investments is difficult to assess.

• More generally, individual investors may have differing attitudes towards the

risks inherent in different classes of investment, further obfuscating the evidence

on rationality of choice.

Finally, the element of time must be considered. A high percentage of UK landlords

have invested in rental property expressly in order to provide themselves a pension (27%

in our sample)—and thus by definition they have a medium to long-term time horizon

(depending on their current age and expected retirement age and preparedness to manage

their portfolio). Monthly or even annual fluctuations in the property market may not

concern them, as long as their cash flow allows them to service their mortgage(s). Even

so, the issue about time horizon is not always straightforward –especially for those with

small portfolios and long time horizons. Active management of the portfolio might lead

dwellings or residential land worth £120,000 - £250,000, 3% on those worth £250,000 - £500,000, and 4%

on those worth over £500,000.

If an investor puts £100,000 in an index fund, and the stock market goes up 10%, the investor has made a

10% return. But if the investor puts a down payment of £100,000 on a property worth £500,000 and

finances the rest with a mortgage, and the value of the property goes up 10% (or £50,000), the investor has

made a return of 50% on his original investment (less interest on the mortgage). Ball notes that many

landlords are outright owners or have a low indebtedness ratio; they could increase their returns by taking

out mortgages up to prudent levels. The fact that they do not suggests that they are financially

unsophisticated and are not maximising returns. (Ball 2004)

to short-term adjustments in holdings even though the long-term objective is clearly to

hold property for liquidation in older age.

4 Factors affecting the rate of return on rental property

If the return on rental property is defined as net rental yield plus change in capital value, a

change in one or more factors could affect the expected return:

• House prices: A reduction in house prices would reduce the capital gain element

of expected rate of return, while increasing the rental yield on open market value;

expectations of increasing property prices would have the opposite effect.

• Rental demand: A fall in demand for rented property could result in lower rents

and higher vacancy levels (both of which would lead to lower net yields). Falling

rent levels or voids can both lead to cash-flow problems for landlords who depend

on rental income to pay the mortgage; landlords without a financial cushion may

be forced to sell. Demand could fall for a number of reasons, including an upturn

in first-time buyer purchases if housing became more affordable through price

reductions, or a decline in immigration as a result of policy or indeed economic

stagnation. Evidence collected by the Investment Property Databank suggests

that UK rental returns have fallen quite significantly over the last two years (IPD

2005). Conversely, an increase in demand for rental property could lead to higher

rents and higher yields;

• Taxes: Any increase in the rate of tax on rental income or capital gains (or

changes in regulations that had an equivalent effect) would reduce the net return

on rental property; decreases in taxes would increase it

;

• Interest rates: A rise in the interest rate would reduce the net yield on rental

property. The average interest rate on a buy-to-let mortgage in January 2006

One implicit tax increase occurred recently: the UK Treasury announced in 2003 that, starting from April

2006, individuals could hold real property through a Self-Invested Personal Pension (SIPP), which would

allow them to avoid capital gains tax on its sale. However, in December 2005 British Chancellor Gordon

Brown announced that he had reversed this decision, and real property would not be permitted in the new

SIPPs. Anecdotal evidence suggests that many investors had already bought, or contracted to buy, rental

units to hold in their SIPPs, and will henceforth face higher than expected taxes when they are eventually

sold.

2005 was about 5%. Because most mortgages in the UK are on variable interest

rates or short-term fixes, most borrowers face higher mortgage payments soon

after rates rise. An increase in interest rates would mean for some landlords that

rents no longer covered their monthly costs. A fall in interest rates would

increase the net yield on rental property.

These factors, and the predicted effect on landlord behaviour, are summarised in Table 1:

Table 1: Factors that might change the expected return on rental property

1: Factors that increase return 2: Factors that decrease return

Rental yield

Lower interest rates

Rising demand for rental property

Decreased taxes on rental income

Falling house prices

Higher interest rates

Falling demand for rental property

Increased taxes on rental income

Rising house prices

Capital value

Rising house prices

Decreased taxes on capital gains

Falling house prices

Increased taxes on capital gains

In general, we would expect a rational investor with a short- to medium-term investment

horizon or an active portfolio management strategy, faced with one or more of the

changes in Column 1, with all else being equal, to purchase more rental property – or at

least not to sell existing holdings. Faced with one or more of the changes in column 2, the

investor would sell rental property – or at least not buy more.

Note that “Falling house prices” and “Rising house prices” appear in both columns.

Falling house prices diminish the capital value of existing portfolios but increase the

potential rental yield on new properties acquired. Expectations about the size and

duration of price falls will be key, and individual investors must assess the net effect of a

change in house prices on the total rental return for their own (existing and potential)

portfolio, in light of their investment objectives. Some will sell on falling prices, and

some will buy—and either decision may be rational. In addition, investors with long-

term investment horizons may think falling house prices herald a buying opportunity,

whatever they mean for returns in the short term.

Alternative opportunities

In addition, improvements in the performance of alternative investments, notably shares,

could cause landlords to disinvest in property; conversely, a fall in the stock market could

lead to increased investment in rental property.

5 Empirical research into landlords’ behaviour in the UK and Denmark

Some research over the last few decades has tried to capture the rich variety of

motivations for being a landlord, and has attempted to identify what this might mean for

landlords’ behaviour as investors. Table 2 summarises the categorization of landlords

from three studies in the UK and Denmark.

John Allen and Linda McDowell, writing in 1989, categorized British private landlords in

the early 1980s in terms of their motives, and the type and amount of property owned.

They concluded that landlords could be split into seven categories (see Table 1), about

half of which could be expected to behave as profit maximisers. (Their categorization

reflects the situation that obtained before the liberalization of the private rental sector in

the UK--in fact, the book sets out partly to explain the disinvestment in private rental

housing--and the categories would be somewhat different today. In particular, the sorts

of landlords covered by the 2004 CML survey analysed below are almost entirely absent

from their categories.)

Allen and MacDowell surveyed a total of 532 privately renting households in the London

boroughs of Hackney and Islington. Most were tenants of investor landlords or informal

resident landlords (that is, they were renting a room or rooms in someone else’s house).

Allen and MacDowell point out that both these types of landlords often had motives other

than strictly economic ones. They estimated, based on a survey of landlords in the

London boroughs of Hackney and Islington, that about half to three-quarters of tenants

lived in properties owned by landlords governed by economic criteria.

Allen & McDowell

categories 1989

Return

maximisers?

PRS H/I* Skifter Anderson

categories 1998

Return

maximisers?

ODPM categories

2003

Return

maximisers?

% dwellings

owned

Traditional, e.g.

Church, charitable

trusts, Crown

Estate

No—service

ideology

0.5/1.6 Public utility: non-profit

companies, private

funds, charities.

No—service

ideology

Employer No—provide

housing for

employees

9.8/4.3 Other companies: Public

companies whose main

business is other than

property ownership

Institutional

landlords—small

% of income from

renting, dwellings

viewed as

investments

Sideline non-

investors

Informal: resident

landlords

small non-resident

No—personal

relationship

Some

39.1/29.1

1.3/5.0

Informal: Individuals

owning one property

with <7 dwellings

Some

Investor: Mainly

longtime owners

who have inherited

properties

Yes—but

constrained by

lack of finance

33.6/42.0

Small investors:

Individuals owning <4

properties but >6

dwellings

Some

Financial: Public

property companies

and pension trust

funds

Yes -/2.6 Financial investors: Pension

funds, banks, insurance

companies

Yes

Sideline investors:

minority of

income fm

letting, viewed as

investment

Yes

Commercial:

speculative, view

rented housing as

commodity.

Yes—but

interested in

capital gains

only

15.7/15.2

Speculators: Companies

organized with very

limited liability

Yes—but

interested in

capital gains

only

Professionals: Individuals

with >3 properties &

companies who run

properties as main

business

Yes

Business landlords:

most of income

fm letting, viewed

as investments

Yes

Builders: Building

companies

Yes—but capital

gains only

Table 2: Categories of landlords (bold italics indicate categories that could encompass 2004 survey respondents)

*PRS H/I: % of households in the private rented sector in Hackney & Islington living in accommodation owned by each landlord type

Hans Skifter Andersen, writing in 1998 about Danish landlords of pre-1950 flats, split

them into eight categories, roughly corresponding to the Allen and McDowell categories.

Skifter Andersen was concerned with understanding why some landlords made rational

economic decisions about whether to invest in the maintenance of their buildings and

others did not. He concluded that only in about half of the eight categories could

landlords be expected to do so. “Studies from Denmark and several other

countries…show that private landlords are a very non-homogeneous group with different

motives and behaviour, where only a minority of landlords fit into the expected picture of

the economically rational private landlord.” (Skifter Andersen 1998, p.196) Skifter

Andersen estimated, based on a survey of Danish landlords, that professional landlords

with mainly purely economic motives owned 30-40% of the lettings (versus 50-75% in

Allen & MacDowell’s UK study).

The Office of the Deputy Prime Minister, in its analysis of the 2001 survey of landlords

carried out as part of the English House Condition Survey (ODPM 2003), used a

categorisation developed by Crook and Kemp in 1996. This divided landlords into four

categories on the basis of the percentage of income received from lettings and whether or

not they viewed their properties as an investment. Landlords in two of these categories,

business investors and sideline investors (together owning about 60% of rental dwellings)

could be expected to maximize returns.

It is clear that over time the landlord sector in the UK has become more business-

oriented--the ‘small non-resident landlords’ identified by Allen and MacDowell 1989

accounted for only 2-5% of lettings in their survey area, while ODPM’s more or less

analogous ‘sideline investors’ owned about 60% of rental properties in 2001.

6 Background on the UK private rental sector

The UK private rented sector, which houses about 10% of households, is smaller than

that in most other developed countries—although the share of renting overall, including

social rental, is similar. In many other respects, though, the private rented sector in the

UK resembles that of other countries. It is a minority tenure--in the UK, as in most

developed countries, owner-occupation is the predominant tenure. Chart 1 shows the

tenure shares in the three countries discussed in this paper, and the US.

Chart 1: Housing tenure in selected countries (2000)

Private renting Owner occupation Social renting Other

USA Australia Denmark UK

Source: Scanlon & Whitehead 2004b

In the UK, as in Australia and Denmark, most landlords are private individuals rather

than companies; they often had personal rather than financial reasons for acquiring

property, such as through inheritance (Ball, 2004; Beer, 1999; Skifter Andersen, 1998;

ODPM 2003). Private rented housing in the UK is not subject to rent control, unlike that

of many European countries, and landlords therefore do not receive the tax breaks

designed to allow them to operate in a rent-controlled environment. (Ball, 2004)

Over most of the last century the private rented sector in the UK was subject to heavy

regulation with respect to both rents and security of tenure. At the same time, private

renting was disadvantaged as compared to owner-occupation in terms of taxation and

subsidy. As a result of these and other pressures (notably slum clearance), the size of the

private rented sector fell to around 9% of the total stock. There was very little

institutional or company investment in housing, and British building societies virtually

stopped lending to landlords after 1945. Individuals tended to be landlords for

idiosyncratic and short-term reasons – they had inherited a property; they needed rental

income to contribute to mortgage costs; they were providing for employees who had to

live near their work; or they were caught in the system by the impact of past regulation

(Nevitt, 1966; Harloe, 1985; Todd, 1986; Kleinman & Whitehead, 1987).

The position changed after 1988 as a result of both supply-side and demand-side changes.

On the supply side, the 1988 Housing Act introduced the Assured Shorthold Tenancy,

which allowed for a minimum rental period of six months, after which a landlord could

give two months’ notice to quit. The Act also abolished regulated rents for new lets,

removing another disincentive for landlords to invest in rental property. In late 1993 a

further reform of the law introduced an accelerated possession procedure, which made it

easier for landlords with a clear right to possession to evict tenants.

These regulatory developments coincided with some significant social and demographic

changes, which contributed on the demand side to the rise of the private rented sector.

These included growth in the student population, an increase in divorce and relationship

breakdown and increasing inward international migration (particularly to London). Over

the same period the relative tax benefits of owner-occupation were reduced by the

reduction and ultimate removal of mortgage tax relief and some relatively minor

improvements in the tax position of landlords.

These factors stimulated a relatively slow growth of the private rental sector over the first

few years, with an estimated 350,000 units added to the private rented sector in England

and Wales between 1989 and 1996 (Kleinman & Whitehead, 1996). Since then, and

particularly since 1999, there has been a more rapid increase; the private rented sector in

the UK now contains some 2.7m dwellings, or about 10.3% of the housing stock (ODPM

live tables), and 12% of households lived in private rented accommodation as of the 2001

census. This increase has been largely funded through dedicated ‘buy-to-let’ mortgages

aimed specifically at landlords, with interest rates only slightly higher than rates on

mortgages for owner-occupied properties. This has brought in a new range of individuals

seeking to invest in property to let privately. Most major mortgage lenders now offer

buy-to-let loans; this has proved an attractive new market for lenders because of the

relatively low risk and the potential for securitisation. As of June 2005 there were

632,100 buy-to-let mortgages outstanding in the UK, according to the Council of

Mortgage Lenders. If every mortgage covers one dwelling, buy-to-let mortgages now

account for approximately 24% of private rented sector.

Simply making funds available, however, was not of itself enough to create demand.

This depended on changing the relative expected rates of return as well as on more

broadly based changes in attitudes. In this context a number of important elements came

together to make investing in residential property more desirable.

First, the upturn in house prices that started in the mid 1990s gathered pace, particularly

in London and the South East. Nationally, house prices rose by 7% p.a. from 1996 –

2000, and in London, where prices were rising most rapidly, they increased by 14% p.a.

This provided an incentive for individuals to invest in residential property on the

expectation of continued future rises. Moreover, as rents could at least be expected to

cover costs, the downside risks were limited.

Second, the stock market experienced a period of heavy decline beginning in early 1994,

but was still thought by many commentators to be overvalued. This provided an

important incentive to shift to what was perceived to be a lower risk investment.

Third, partly as a result of this decline, many pension funds ran into trouble and the

expected value of many pensions was reduced. For individuals wanting to make their

own arrangements for retirement income, property appeared to offer a safe bet. Fourth,

over the period interest rates have fallen to historically low nominal levels. In 1998, the

average interest rate on a UK mortgage was 7.19%; by 2004 this had fallen to 5.18%.

This lessened landlords’ outgoings and therefore further reduced the risks of investment.

As of 2001, about 65% of private rented sector dwellings were owned by individual

landlords, and about 30% by companies or institutions (ODPM 2003).

The British government has not viewed the rapid growth in individual investment in the

private rented sector as an unalloyed good; it has made clear that it wants to encourage

increased investment in the sector by institutions such as pension funds. An increase in

institutional investment “could allow more efficient management, greater renewal and

preservation of property, and a more liquid supply of housing…This potential in the UK

for greater corporate and institutional investment in the PRS should also raise standards

and provide an alternative to the highly geared buy-to-let market.” (HM Treasury, 2004,

p.12)

7 The 2004 survey of UK buy-to-let landlords

Do landlords in fact make economically rational decisions? We define rational behaviour

as that which is directed towards maximization of returns. Faced with economic factors

that decreased the relative return on rented housing (see Table 1), a rational landlord

would choose NOT to increase holdings of such property. Faced with economic factors

that enhanced the relative return on rented housing, a rational landlord would chose NOT

to decrease holdings of such property.

Our hypothesis is that landlords behave in a way that is consistent with economic

rationality. Data from a recent survey of UK landlords allows us to test this. In 2004, the

Council for Mortgage Lenders (CML) commissioned the largest survey to date of buy-to-

let investors in the UK. The CML has generously allowed us to use the data for this

analysis. The questionnaire was designed to elicit information for CML’s own use, and

The domination of the PRS by small landlords is viewed as a problem in other countries as well; Berry

and Hall (2005) argue that ‘the absence of large investors has, arguably, intensified the housing

affordability crisis…in Australia.”

thus some questions were not worded in the way that would have best suited our

purposes.

Twelve mortgage lenders active in the buy-to-let market sent postal questionnaires to 700

randomly chosen buy-to-let customers (8400 in total). The overall response rate was

15.8% (1341 responses). This survey does not represent the whole spectrum of

residential landlords, in particular because it excludes those landlords who have no

mortgages on their rental property. Research by the ODPM in 2001 estimated that 31%

of landlords had a loan or mortgage outstanding on their properties (ODPM 2001).

Crook and Kemp found that in 1993-94, about 45% of lettings had been bought with

cash. The Mintel survey found that 53% of landlords purchased their property without

using a mortgage. By definition the survey excludes landlords in several of the other

researchers’ categories, notably traditional/public utility/institutional landlords;

employers; and financial institutions. The names in bold italics in Table 2 are those

landlord categories that might encompass the respondents to the 2004 CML survey.

We would expect the surveyed landlords to be more responsive to economic stimuli than

landlords as a whole, because they all have mortgages on at least some of their properties,

and in most cases bought those properties specifically in order to rent them out. A brief

general description of the sample: 27% of the landlords surveyed owned one property

and 16% owned two. Two percent of landlords surveyed owned more than 50 properties.

73% had been landlords for at least two years, and 18.5% for more than ten years. More

than half the landlords surveyed owned some properties that were not backed by a

mortgage (owned outright). (For a fuller report on the survey findings, see Scanlon &

Whitehead 2004a.)

The overwhelming majority said their main reason for buying rental property in the first

place was broadly financial. 31% said they would ‘rather invest in property than other

investments; 26% saw it as a contribution to pension provision, and 19% were seeking

‘return on investment: growth plus rental income’. (Clearly these goals are not mutually

exclusive.)

8 Categorisation of buy-to-let landlords

The criteria used by other researchers in assigning landlords to categories are set out in

Table 3. The last two columns indicate whether each criterion seemed, a priori, a useful

predictor of the economic rationality of the landlord, and whether the CML survey

contained data on it.

Table 3: Criteria used in assigning landlords to categories

Criterion Researcher Predictor of ec

rationality?

Data available?

Type of owner (individual;

church/charity/Crown;

employer; financial

business)

Allen &

McDowell;

Skifter

Andersen

Yes—at least

insofar as some do

NOT maximize

returns (e.g.

charities)

No--Charities,

employers not

included in our

sample.

Means of acquisition of

property

(inherited/purchased)

Allen &

McDowell

Possibly No--all ours own

at least some that

they purchased

Whether landlord is resident

or nonresident

Allen &

McDowell

Probably No—all ours are

nonresident

Whether landlord has

speculative motives

Allen &

McDowell;

Skifter

Andersen

Possibly No—survey does

not address

motivation in this

way

Number of dwellings and

properties owned

Skifter

Andersen

Possibly Yes

Whether property is main

business/% of income

derived from letting

Skifter

Andersen;

ODPM

Yes Yes

Whether properties viewed

as investments

ODPM

Yes Yes

The CML data did contain data on most of the criteria we felt were definitely relevant to

the economic rationality of landlords. Those criteria we considered possible predictors of

economic rationality, and for which we had data, were:

• Number of properties owned. Those landlords owning many properties, with a

commensurate amount of money tied up in them, could be expected to make more

economically rational decisions than those owning only one or two.

• Whether property is main business/% of income derived from letting. We expect

landlords who consider rental property to be their main business, or for whom

rental income accounts for a high percentage of income, to be more governed by

economic factors than landlords for whom renting is a minor sideline.

• Whether properties are viewed as investments. Landlords who explicitly state

that they have investment goals for their rental units should make more

economically rational decisions than those who own rental property for other

reasons.

We divided the sample into two sets, which we call Professional and Non-professional

landlords. We define Professionals as those landlords who owned at least three

properties; whose letting income was at least 50% of their total income; and who had

financial, business or pension motives for being a landlord. Thus defined, Professionals

made up 15% of the sample. Non-professionals were all other landlords. We expected

that Professionals would exhibit a more commercial—or more economically rational--

attitude to the acquisition and management of their portfolio than Non-professionals.

9 Analysis of economic rationality

The survey asked landlords several questions about how they might react to different

economic scenarios. They were asked what might cause them to increase their portfolio

of rental properties. Table 4 summarises the response from the whole sample, and from

Professional and Non-professional landlords.

The wording of the questionnaire was as follows: “How do you currently view your primary role as a

residential landlord? 1. Am running full-time business; 2. Investment for capital growth only; 3. Investment

for rental income only; 4. Investment for capital growth and rental income; 5. Provides somewhere for

me/my family to live; 6. Temporary because cannot sell property; 7. Contribution to pension provision; 8.

House an employee; 9. Other.” Answers 5, 6 and 8 were considered to show non-investment motives.

Table 4: Reasons to increase portfolio

Percentage supporting* Predicted reasons to

increase portfolio (fm

Table 1)

Wording of survey

All Profs Non-

profs

Lower interest rates

Stable/low interest rates 57

63 53

Rising house prices

Steady/rising house prices 48 51 45

Rising demand for rental

property

Very good rental yields 45 58 40

Worsening performance of

alternative investments

Poor stock market

performance

21 20 20

Help children/family into

housing market

15 10 15

Improved job

prospects/bonuses

9 2 10

Children at university 9 7 9

Other 10 10 10

*Percentages do not add to 100 as multiple responses were possible

The answers given broadly agreed with those predicted in Table 1. Landlords were far

more likely to cite financial than personal reasons for increasing their portfolios. We

would not necessarily expect all participants in a market to be equally affected by

particular stimuli, and indeed even the most-frequently given reason for increasing a

portfolio—stable or low interest rates—only attracted the support of 57% of respondents,

and none of the others garnered more than 50%. Factors related to the performance of

rental housing itself as an investment were seen as far more important than the

performance of alternative investments; ‘poor stock market performance’ was cited by

only 21% of respondents as a reason to increase rental portfolios.

The second most-cited reason given by landlords for increasing their portfolios was

‘rising house prices.’ This suggests that landlords are more interested in capital gains

Among the factors in Table 1 affecting rate of return were changes in tax rates and falling house prices.

They are not, however, included in Tables 3 and 4, as the survey did not ask about them.

(which would rise with rising house prices) than in rental yield (which would fall as

house prices rose).

Professional landlords were more influenced than Non-professionals by cash-flow

considerations: they were more responsive to interest-rate changes and rental yields.

The survey also asked landlords what might cause them to reduce their portfolios of

rental property. Responses are summarized in Table 5.

Table 5: Reasons to decrease portfolio

Percentage supporting* Predicted reasons to

decrease portfolio (fm

Table 1)

Wording of survey

All Profs Non-

profs

Falling demand for

rental property

Rental income consistently

not paying off mortgages

Higher interest rates

Rising interest rates 47 58 45

Worsening personal financial

situation

26 8 30

Approaching retirement 19 20 19

Falling house prices over 3-6

months**

16 14

Falling house prices

Stagnating house prices: 0%

growth over 3-6months**

7 7 6

Worsening general economic

situation

8 14 8

Improved performance

of alternative

investments

Improving stock market 4 3 4

Children leaving home 2 2 2

*Percentages do not add to 100 as multiple responses were possible

**16% of all respondents said either falling or stagnating house prices over 3-6 months might cause them

to decrease their portfolios.

Here as well, the two reasons landlords gave most frequently for decreasing their

portfolios were financial ones, broadly in accordance with predictions. The next most

commonly cited reasons, ‘worsening personal financial situation’ and ‘approaching

retirement’, could be expected to lead to reductions in an investor’s overall portfolio,

including rental property.

As in Table 4, the performance of rental property itself as an investment was

much more important than the performance of alternative investments. Only 4%

of landlords said an improving stock market would be a reason to decrease their

portfolios (compared to 20% who agreed with the converse).

While 48% of landlords said that stable or rising property prices were a reason to increase

their portfolio, only 16% supported the converse: that falling or stagnating property

prices are a reason to sell. This may be due in part to the time frame of 3-6 months given

for property price falls in the questionnaire (no analogous time frame was given for price

increases), which is short compared to the investment horizon of many landlords.

Both the relative unimportance attached to returns in alternative markets, and the weak

response to falling property prices, may in part be consequences of the significant

financial hurdle represented by transactions costs for real property: very large price

movements (for stocks or property) could be required to offset the transactions costs

involved in making changes to portfolio of rental property.

Table 5 also highlights the importance of cash flow to landlords. The single most

important reason landlords gave for reducing their holdings of rental property was ‘rental

income consistently not paying off mortgages’, cited by 49% of all respondents. Rent not

covering the mortgage could be a consequence of either or both of two factors that lead to

lower returns: rising interest rates and/or a fall in demand for rental housing (which

could lead to void periods as well as lower rents). The key here is the nexus between

rental income and mortgage payments—if for some reason the rent did not cover

mortgage payments, many landlords would have no choice but to sell, whatever they

might think about the future of property as an investment.

Professional landlords were again more likely than Non-professionals to say they would

react to changes in interest rates, and much less likely to say that changes in their

personal financial situation would affect their property holdings.

The questionnaire asked landlords what they would do in the face of some scenarios.

One was as follows: ‘If, in the next 12 months, interest rates rise to 5.75% (currently

4.75) and were expected to stay at this level or rise going forwards, how would this affect

your role in residential letting?’ The results are given in Table 6.

Table 6: Predicted reaction if interest rates rose by 1%

Percentage Reaction to increased interest rates

All Profs Non-profs

No effect 62 50 65

Trade/rebalance portfolio but keep investment 14 25 13

Sell some/all of portfolio 14 17 13

More likely to increase portfolio 5 10 5

Leave the market altogether 4 1 4

Most landlords said that a 1% increase in interest rates would not affect their portfolios.

Of those who said it would have an effect, three times as many said they would sell

properties as buy them, which is consistent with the predictions in Table 1.

Professionals were more likely than Non-professionals to say that they would react to a

rise in interest rates. One-quarter said they would trade or rebalance their portfolio but

keep their investment, while 17% said they would sell some of their portfolio and 10%

said they would increase their holdings.

Buying more property after interest rates rise is inconsistent with the predictions in Table

1, yet Professionals (whom we expect to be more commercially minded) were twice as

likely as Non-professionals to say that they would do so. This may simply mean that

Professionals could still extract profit in a higher-interest-rate environment and so would

continue to add to their portfolio regardless, although their additional investment might

be smaller in scale or more selective than it would be if interest rates were unchanged.

Alternatively, Professionals might assume that higher interest rates would trigger price

falls, which could offer compensating increases in gross rental yield.

At the same time, paradoxically, Professional landlords were more likely than Non-

professionals to say that rising interest rates would be a reason to sell properties (Table

5). The higher propensity of Professionals to buy or sell with a change in interest rates

suggests they may have less emotional attachment to individual properties than Non-

professionals, and would buy/sell/rebalance to maximize their return.

10 Conclusions and policy implications

Our conclusions are necessarily rather broad-brush, not least because the UK buy-to-let

sector is still so young. Many investors only have experience of a rising property market-

-and a survey about hypothetical situations has obvious limits. So faced with, say, a fall

in property prices, landlords may in fact do something very different from what they said

they would do. In addition, it is difficult to design a simple test to determine whether

investors’ decisions are ‘rational’ or not. A decision that seems irrational or even

perverse, given one set of assumptions about investment horizons and expectations, may

seem perfectly reasonable under another set of assumptions. Even so, some broad

conclusions do emerge from the data.

First, most landlords are sensitive to the rate of return on rental housing, and when faced

with various housing-market stimuli, say they would act in a way that is not inconsistent

with maximizing return. A significant minority say they would behave in a way that

would seem, a priori, to be inconsistent with return maximization. However, without

more detailed information about their preferences, it is too simple to characterize these

choices as ‘irrational.’ In particular, investors’ different time horizons and investment

approaches can lead to different outcomes: a short-term, speculative landlord

concentrating on capital gains may buy into a rising market, on negative rental yields, in

the expectation of selling the following year; while someone buying property as a

pension may buy during a market downturn, happy to buy cheaply an asset that will be

held for 20 years.

Second, while most buy-to-let landlords are sensitive to the rate of return on rental

housing, most do not appear to be directly sensitive to the rate of return relative to other

investments, particularly the stock market. This does not accord with the standard view

of investor behaviour, in which investors seek to maximize returns. This may be in part

because it is very difficult, even for specialists, to make accurate comparisons of returns

in the two markets—particularly given the fact that rental property is often financed with

borrowed money. It may also reflect landlords’ long investment horizon: 63% of those

surveyed said they planned to remain in the residential rental market for ten years or

more, and 27% said they viewed their properties as part of their pensions. However,

some 20% of landlords said falling stock prices were a reason to buy property, but only

one fifth as many--4%--said rising stock prices were a reason to sell property. This

suggests that they do not assess their overall investment portfolios objectively and

dispassionately, but rather have a general bias in favour of buying rather than selling

property. On the whole, therefore, our findings seem to support Shiller’s observation that

investors do not, in fact, evaluate the entire universe of investment possibilities in making

their decisions.

Given landlords’ general insensitivity to developments in other markets, and the

transactions costs problems, it is perhaps more useful to think of landlords as small

businesspeople rather than straightforward investors. US-style REITs, which allow for

passive, collective investment in rental properties, are an investment that can be

compared in like terms with other investments; residential rental properties are not.

Third, although these buy-to-let landlords tend to claim to have long time horizons for

their investments, about half of them say that rising interest rates might be a reason to sell

properties. Indeed, when asked about a specific interest rate rise—1% over a year—19%

said they would sell some or all of their properties. This is still a sizeable percentage.

This suggests that short-term considerations, particularly with respect to cash flow, would

trump expectations of long-term capital gains.

Fourth, Professionals are more likely to say they would react to changes in two important

economic factors, house prices and interest rates, than Non-professionals – which does

point to different approaches to assessment and decision dependent on scale and attitude.

It does not of course necessarily point to different outcomes.

Our conclusions echo those of Andrew Beer’s 1999 study of private-sector landlords in

Australia:

First, the ability of private rental housing to offer a secure long-term investment is

the dominant motive for investment. Second, landlords are drawn from all

income groups, Third, individuals invest in rental housing for disparate

reasons…there is therefore a degree of unpredictability within the private rental

market, not all investors are likely to respond quickly to changes in investment

regimes or market conditions and this is particularly true for those on lower

incomes. (Beer, 1999, p.260)

Beer found that landlord income did affect economic rationality. Unfortunately our data

did not allow us to test this explicitly, but the average income of the Professional

landlords in our survey (who were more responsive to market conditions) was almost

certainly higher than that of the Non-professionals.

What implications do these findings have for policy? Changes over the last 20 years in

the policy and financial landscape in the UK have led to an upsurge in the number of

private landlords, in line with government intentions. The government is less pleased

that this increase has come almost entirely from the ranks of private individuals. It would

like to change the mix of private landlords by encouraging corporations and financial

Interestingly, Beer found that over a third of all landlords, and 40% of those with mortgages, did not have

a clear idea of their income position—that is, they did not know whether they were making money or not.

Clearly such landlords cannot be return maximisers, at least in the short term.

institutions to invest in rental property. Although this has been the government’s goal for

some years now, it has failed to make much headway; corporations remain resolutely

uninterested in becoming residential landlords. Our findings indicate that individual

private landlords do generally respond to economic stimuli in rational ways. While

Professional landlords are more responsive to some stimuli—particularly interest-rate

changes--than Non-professionals, the difference is not enormous. It may be that policy

would be better directed toward helping individual landlords develop financial and

management skills than in continuing to try to attract direct corporate investment in the

private rented sector.

Bibliography

Allen J & McDowell L (1989) Landlords and Property: Social Relations in the Private

Rented Sector Cambridge University Press, Cambridge.

Association of Residential Letting Agents (2005) “The ARLA Review & Index of

Returns on Residential Investment, Third Quarter 2005”, ARLA, London

Baddeley M (2005) “Housing Bubbles, Herds and Frenzies: Evidence from British

Housing Markets” Centre for Economic and Public Policy Brief No. 02/05, Centre for

Economic and Public Policy, University of Cambridge, Cambridge.

Ball M (2004) The Future of Private Renting in the UK Social Market Foundation,

London.

Ball M & Glascock J (2004) Property Investment Funds for the UK Council of Mortgage

Lenders, London.

Beer A (1999) “Housing Investment and the Private Rental Sector in Australia”, Urban

Studies 36(2), 255-269.

Berry M and Hall J (2005) “Institutional Investment in Rental Housing in Australia: A

Policy Framework and Two Models”, Urban Studies 42(1), 91-111.

Clapp J, Goldberg M and Myers D (1994) “Crisis in Methodology: Paradigms and

Practice in Real Estate Research” in Appraisal, Market Analysis & Public Policy in Real

Estate (DeLisle J & and Sa-Aadhu J, editors), Kluwer Academic Publications, Boston.

Crook A and Kemp P (1996) ‘The Revival of Private Rented Housing in Britain’ Housing

Studies 11(1), 51-68.

Crook A, Kemp P, Barnes Y and Ward S (2002) Investment Returns in the Private

Rented Housing Sector, British Property Federation, London.

Harvey J (1981) The economics of real property Macmillan, London.

Harloe M (1985) Private Rented Housing in the US and Europe Croom Helm,

London/Sydney.

Hendershott P and MacGregor A (2006) Investor Rationality: evidence from UK

capitalisation rates, RICS, London.

HM Treasury (2004) Promoting more flexible investment in property: a consultation HM

Treasury, London

Investment Property Databank (2005) “IPD UK Residential Investment Index”.

Available online at

http://www.ipdindex.co.uk/downloads/indices/UKResidential_2005.pdf

Mintel Consultancy (October 2004) The Changing Face of the UK Landlord Charcol

Mortgages, London.

Nordvik V (2001) “Analysis of Rental Housing Markets: Five Essays” Norwegian

Building Research Institute (Byggforsk), Oslo. Summary online at

http://www.byggforsk.no/default.aspx?DokumentID=794&innholdsID=114&spraak=en

Office of the Deputy Prime Minister (2003) English House Condition Survey 2001:

Private Landlords Survey ODPM, London

Office of the Deputy Prime Minister (2006), live tables on housing stock, online at

http://www.odpm.gov.uk/index.asp?id=1156006

Paley B (1978) Attitudes to letting in 1976 Office of Population Censuses & Surveys,

HMSO, London.

Poterba J (1984) “Tax subsidies to owner-occupied housing: an asset-market approach”

Quarterly Journal of Economics 99:729-752

Rhodes D & Bevan M (2003) Private landlords and buy-to-let Centre for Housing

Policy, University of York, York.

Scanlon K & Whitehead C (2004a) The profile and intentions of buy-to-let investors

Council of Mortgage Lenders, London.

Scanlon K & Whitehead C (2004b) International trends in housing tenure and mortgage

finance Council of Mortgage Lenders, London.

Shiller R J (2000) Irrational Exuberance Princeton University Press, Princeton.

Shiller R J (2003) “From Efficient Markets Theory to Behavioural Finance” Journal of

Economic Perspectives 17(1) 83-104.

Skifter Andersen, Hans (1998) “Motives for Investments in Housing Rehabilitation

among Private Landlords under Rent Control” Housing Studies 13(2) 177-200.

Thaler R. (1980) “Towards a positive theory of consumer choice” Journal of Economic

Behavior and Organization 1, 39-60.

Todd J E (1986) Recent Private Lettings HMSO, London.

Vass J (2004) “Buy-to-let: long-term stability or short-term volatility?” Market Briefing

Special Article, Council of Mortgage Lenders, London