Knowing the Area: The Management of Market and Business Risks by Private Landlords in Scotland

Abstract

Compared with the management of social rented housing in Britain, comparatively little is known about how private landlords manage their property, despite their increasing importance. This paper reports results from a recent survey in Scotland which shows that landlords restrict the acquisition of property close to where they themselves live to manage the market and business risks they face. This is because their personal knowledge of markets is important in reducing risks, given the complexities of the sub-markets where they operate and the information asymmetries involved. The paper combines quantitative data on a large sample of properties in Scotland and their landlords, including the location of their properties and homes (or businesses), with a series of focus groups with a smaller sample to understand the reasons why most landlords live very close to their portfolios. The policy implications of the results of the surveys are also discussed.

1. Introduction

Governments in Britain, both in Westminster and in the devolved administrations, have increasingly emphasised the role that the private rented sector might play in meeting contemporary housing need. Despite a range of policies over the past two decades aimed at professionalising the sector and securing greater institutional investment, it remains largely a ‘cottage industry’ composed mainly of individual landlords with small portfolios, many managing property in their spare time (Crook and Kemp, 2010; Rugg and Rhodes, 2008). The business of private letting mostly involves active, rather than passive, forms of investment and therefore an understanding of the associated risks.

Like all businesses, small landlords must confront both market and business risks. The basic tasks involving market risk are twofold: buying properties and setting rents. Having invested, landlords have little control over the capital values of their properties and the rents they can earn, and hence need to find ways of managing these risks as best they can at the point of investing. This takes place in a market where the information required to decide what to pay or when to sell and at what price is not easily obtainable. Nor is information on what rents to charge easily obtainable, especially given that landlords are primarily rent ‘takers’, setting their rents by reference to the prevailing local market rate (Kemp and Rhodes, 1997).

Alongside market risks, landlords’ day-to-day management tasks involve dealing with an array of business risks. Landlords seek to minimise these by making decisions on letting policies, finding suitable tenants, minimising voids between tenancies, ensuring rents are paid on time, minimising arrears, chasing bad debts, ensuring that urgent and longer-term repairs across their portfolio are identified and carried out, and managing tenants’ anti-social behaviour. While they have more control over these business risks than they do over market risks, landlords still need to find ways of minimising them and of securing the information they need to inform their activities and decisions.

This paper attempts to understand how private landlords manage market and business risks given their characteristics and the nature of the local markets in which they operate. It seeks to address a hitherto underdeveloped understanding of the role of the geographical relations between landlords and their portfolios. We develop this understanding within an informational economics framework where information on local sub-market conditions is essential to the minimisation of the two types of risks that we have described. We draw on the results of a recent survey of private landlords in Scotland to examine these issues.

The paper has six further sections. A short review of the literature on sub-markets and information economics, which is used as a theoretical framework, is followed by a brief discussion of the relevant policy background. An explanation of our methods and data is then given. We then present an analysis of our findings structured into two sections: first, the patterns of ownership and, secondly, the management of risks in the Scottish private rented sector. We conclude by discussing the implications of our findings for policy and future research.

2. Sub-markets and Information

That housing markets do not operate perfectly is well known. In particular, housing is a highly heterogeneous and illiquid good with fixed locations (Maclennan, 1982). Buying and renting decisions are taken more rarely than the purchase of other goods, while local information about dwellings, including prices, is hard to come by without substantial search costs. Wheaton (1990) demonstrated that the housing market can be conceptualised as a ‘search market’ in which the actual activity of seeking out sale and purchase opportunities is both defining of the market and contributory to the structure of its outcomes. The quality of information on the market and the ease with which this information can be collected and assimilated are important components of overall search costs.

A great deal of research on the nature of housing markets has also stressed the need to disaggregate the market analytically into a series of spatially differentiated, although interrelated, sub-markets (Grigsby, 1963; Galster, 1996; Maclennan and Tu, 1996; Watkins, 2008). It has been suggested that such disaggregation is especially important for rental markets (Allen et al., 1995). Sub-market theory suggests that similar properties in different, but not dissimilar neighbourhoods, will not necessarily have the same prices and also that the structure of sub-markets tends to be relatively rigid and persist over quite lengthy periods (Maclennan and Tu, 1996; Jones et al., 2002, 2003).

Sub-markets are not entirely spatial phenomena, however. Markets can also be disaggregated in terms of property and tenure characteristics, emphasising the importance of the substitutability of individual properties for each other. The substitutability concept is applicable equally to prospective tenants’ search behaviour and landlords’ market decisions about property acquisition/disposal, although the latter is more immediately relevant to this paper. Both spatial and aspatial sub-market definitions place a premium on the value of information to understand how sub-markets are composed and how prices and rents are formed within them.

The smooth movements of households and flows of transactions and information between sub-markets may be disrupted by high search costs, information inefficiencies and other barriers (Maclennan and Tu, 1996) including institutionally created supply inelasticities such as planning constraints and the fact that households’ satisfaction levels with their existing neighbourhoods result in a tendency to move within them rather than to other neighbourhoods. This means that it may be very hard to acquire relevant information about all sub-markets within, say, a large urban area and that the interaction of supply and demand, and hence prices (including rents), may well take place within quite small areas or sets of substitutable properties rather than across the series of sub-markets. The existence of sub-markets frustrates the process of spatial arbitrage and thus impedes the optimal clearing of the unitary market (Meen, 2001; Jones et al., 2003). Knowing what the market rent in one neighbourhood is for, say, a two-bedroom flat may tell little about prices or rents for similar dwellings in another neighbourhood. Thus excess supply and excess demand can be observed to co-exist over time across the various sub-markets making up a metropolitan area (Jones et al., 2004; Jones and MacDonald, 2004).

Information economics shows that information is often hard for individuals to obtain (Maclennan, 1982), especially in highly differentiated or spatially segmented markets where there are many buyers and sellers. We further know that information asymmetries in markets often mean that sellers or landlords have very different information sets from those of potential buyers or tenants (Stigler, 1961; Read, 1997) and that there exist significant informational uncertainties (Yavas, 2001). Although estate agents and letting agents have the function of bringing suppliers and demanders together and improving informational flows, this depends on them being effective and trusted. In the UK, there is no statutory regulation of managing agents and the adoption of voluntary codes is minimal (Crook and Kemp, 2010; Law Commission, 2007). Moreover, there is a body of work that suggests that information asymmetries are the basis for negotiations and can be actively exploited by skilful buyers to acquire products below market prices. Recent research on US commercial real estate markets suggests that information asymmetries are important in property markets; that those taking part deliberately buy properties in nearby locations; and that they also avoid dealing with professional agents known to be well informed (Garmaise and Moskowitz, 2004). It has also been shown that agents selling properties that they themselves own can command a premium price (Rutherford et al., 2007). Wallace (2008) has shown how important information gained informally is in the making of local housing markets and Hickman et al.’s (2008) study of private landlords also emphasises the local focus of many landlords’ operations and the central role of local knowledge in informing their behaviour. Yet, beyond some studies commenting on the changing geography of rental markets (Gibb, 1994; Gibb and Nygaard, 2004), little is known about the specific geographical relationship between individual landlords and their portfolios.

These concepts of local information and geographical proximity are useful in understanding how private landlords manage market risks. It is hard for the small-area, part-time landlord characteristic of landlords in Britain and most other advanced economies (Priemus and Maclennan, 1998) operating within a large metropolitan area to acquire all the information needed to buy and manage properties in ways that minimise risk. Among other things, it requires the landlord to know if the price being paid is good value and what rents to charge. There is evidence to suggest that landlords do not always seek to maximise rents but may instead adopt ‘turnover minimising’ strategies to minimise vacancies, either by letting at below market rents (Bevan et al., 1995) or by not raising rents at the end of a tenancy (Kemp, 2004).

With some notable exceptions (for example, Thomas and Snape, 1995; Bevan et al. 1995; Kemp, 2004), comparatively little is known about how contemporary landlords manage business risks within their businesses except that many do it entirely by themselves in their spare time. The costs and risks of managing residential lettings can be considerable. Work undertaken in England (Crook et al., 2002) shows that the costs of management tasks range from one-third to two-fifths of gross rental income and that poor management can cost landlords 10 per cent of average gross annual rents in voids and arrears.

Implicitly, then, there is a trade-off between managing business risks and maximising rental yields. Kemp (2004) notes that landlords seeking to minimise turnover also incur lower business costs and that it is hence possible that turnover minimisers’ net yields are not necessarily much lower than those of rent maximisers. The efficient pursuit of either strategy requires landlords to have good information about market rents in their localities of operation. Although agents’ valuations of properties should in theory help to mitigate market risk and pursue an efficient strategy, the volatility of the UK residential property market and the dominance of owner-occupation makes discerning value against potential rental yields more difficult. Residential rental indices remain poorly developed, in contrast to indices of residential sale prices and commercial property rents (Rhodes and Kemp, 2002). If landlords had access to good and trusted letting agents, the business risks would also be less problematic. Yet where the services of agents are underregulated, not available or otherwise underused, it may be reasonable to expect that landlords may restrict their operations to areas they know well.

The existence of sub-markets and information asymmetries in rental property markets is of policy significance because these features, alongside the structure of landlords and their portfolios, may impose substantial limits on the way that the private rented sector can develop. This is relevant because there has been increasing interest in Scotland, as elsewhere, in the roles that the sector might play in meeting housing need and fulfilling a range of housing and social policy objectives. We now briefly expand on this point by sketching the policy background to our study.

3. Policy Background

After decades of decline for much of the past century, the private rented sector in Britain is expanding. This has inevitably raised the possibility that a larger, more diverse sector could meet wider housing and social policy objectives. Understanding how private landlords manage their portfolios lies at the heart of any attempt to deliver these wider objectives. There have been three key ‘drivers’ to policy about private rented housing in recent years: the role of the sector, its ownership structure and its management. Each is briefly introduced in turn.

First, attitudes to private rented housing have changed significantly in Britain over the past 30 years and there is now widespread agreement that it has a positive role to play in meeting significant segments of Britain’s housing needs (Best et al., 1992; Crook and Kemp, 1996; Scottish Government, 2007; HM Treasury, 2010). To help achieve an expanded sector, rent deregulation (in 1989) and associated measures were designed to create a framework whereby an entrepreneurial rental market could rapidly grow in size, encouraged by rent-setting freedom and greater liquidity (Lowe, 2007).

Secondly, since the 1980s British governments have been concerned about the structure and financing of the sector. They have wanted to attract more landlords with corporate funding such as pension funds. The view was initially taken that the sector was unlikely to grow substantially if it was reliant only on individual investors’ ability to borrow (or invest their own equity). In contrast, corporate landlords might take a longer-term view of returns on investment, thus creating more stability in the sector (Crook and Kemp, 2002). It was also thought that larger landlords might also develop portfolios in different sub-markets and parts of the country, thereby diversifying market risks. Business risks might also be addressed by realising economies of scale in management and maintenance. Several government schemes aimed at stimulating institutional investment had only limited success. Meanwhile the industry’s own ‘Buy-to-Let’ scheme diversified investment into rental property, mainly from individuals (Gibb and Nygaard, 2004; Leyshon and French, 2009), and the sector remains dominated by individual landlords (Crook and Kemp, 2002, 2010).

Thirdly, there has been continuing concern about housing standards and management practices in the sector. In recent years in Britain, and especially Scotland, more emphasis has been placed on securing better management of deregulated rental housing. This reflects ongoing concerns about poor quality in parts of the sector and also governments’ desires that it should be used to accommodate homeless households through a range of schemes including leasing (Pawson, 2007; Rugg and Rhodes, 2008).

The impact of these three strands of policy has been mixed. The sector grew substantially in size, especially during the house price boom of the early 2000s. In 1991 the private rented sector housed only 6 per cent of Scotland’s households; by 2008, it housed nearly 9 per cent (Scottish Government, 2009). However, significant corporate financing has yet to emerge and there is continuing evidence of poor management standards at the bottom end of the market (Scottish Government, 2009).

While there is general agreement about the overall objective of expanding the sector, views differ as to how this might best be achieved. A recent consultation by the UK Treasury suggested that there remain significant tax disincentives to major investment in rental property. The Coalition government in the UK has been reluctant to reform taxes at a time of fiscal austerity, however, believing instead that an economic upturn and increase in debt finance will in time create a more benign environment for investment. Nevertheless, it has reformed Stamp Duty Land Tax on bulk purchases by placing the tax on the average value of each property acquired and not the entire portfolio (HM Treasury, 2010).

The issue of dwelling quality was addressed in the Rugg review of the sector (Rugg and Rhodes, 2008), which proposed the introduction of licensing for landlords and rental property managers. The Coalition government rejected those proposals, claiming that the current arrangements “strike the right balance between tenants and landlords.” (Hansard, 2010). Given that the current UK government believes that the rental market operates in a way that does not warrant further regulation or incentives, a fuller understanding of the risks facing landlords is required to understand the potential of the sector expanding to meet wider housing policy objectives.

4. Methods and Data

The empirical data presented in this paper are based on a major survey of landlords of a representative sample of privately rented addresses in Scotland. The nearest equivalent of a universal database of privately rented addresses is the Scottish private landlord registration system managed by local councils. The Antisocial Behaviour etc. (Scotland) Act 2004 requires all private landlords, with a few clearly defined exceptions, to register themselves and their properties with the local authorities in which their properties are located. Although there was not full compliance with this requirement at the time of sampling, it provided the best available source from which to construct a sample frame of privately rented properties.

The registration database was sampled using a two-stage process. First, a random sample of 16 local authorities was drawn, representing half of Scotland’s 32 unitary authorities. The sample was constrained to include both Edinburgh and Glasgow as together these cities represent a large part of Scotland’s rental sector. Secondly, consent was sought from these authorities to sample addresses from their registration database. Difficulties in obtaining co-operation reduced the sample of local authorities to 14. These are listed in the Appendix and represent a range of geographical and market contexts.

A systematic sample of every 15th address was drawn from the register. Duplicate landlords were removed from the sample where possible. In total, 5893 addresses were sampled. Questionnaires were sent to the owners of these addresses by both mail and e-mail. Recipients could elect to return a hard copy of the questionnaire or complete it on-line at a dedicated website. Follow-up e-mails and letters were sent three weeks after the initial mailing. When the survey closed in autumn 2008 a response rate of 26 per cent had been achieved, resulting in a dataset of the landlords of 1468 addresses (63 responses were from landlords’ agents).

The achieved sample size enabled confidence in the results of ±2.5 per cent of the mean at the 95 per cent confidence level, assuming variability of 50 per cent for any variable (see Dixon and Leach, 1978). Using information on all properties in the registration database, the data were weighted by local authority so that addresses in authorities of different types (rural, urban, large cities) were represented in their correct proportions. This corrected for the oversampling that resulted from the forced inclusion of Edinburgh and Glasgow. Crook et al. (2009) provide full details of the sample and survey instrument.

In order to explore the survey findings in more depth, focus groups were conducted with a sub-sample of respondents. A total of 90 landlords attended nine focus groups held in different parts of Scotland. Those who attended were a broad cross-section of respondents in terms of portfolio size, landlord type and investment orientation, with a slightly greater attendance by those with an investment orientation and those expecting to expand their portfolios.

Finally, a geographical information system was used to analyse survey responses and the registration database to examine the spatial relationship between landlords and their properties. The postcodes of sampled property addresses and the home or business addresses of their landlords were compared to analyse how close landlords live and work to their portfolios.

5. New Evidence on Landlords in Scotland

We found that in Scotland well over three-quarters of private rented dwellings are now owned by individuals and couples, a proportion that has grown in recent years (see Kemp and Rhodes, 1994, 1997), as it has elsewhere in Britain (Crook and Kemp, 2010). Characterising the sector as a ‘cottage industry’ is as relevant now as ever. Before examining the survey evidence about how landlords perceive and manage risk, it is pertinent to describe further Scotland’s landlords in terms of their type, their investment orientation, their entry into the market, their portfolio size and how their portfolios are managed.

5.1 Landlord type

A previous study in Scotland (Kemp and Rhodes, 1994) reported that in 1992/93 just less than half (47 per cent) of the stock had landlords who were private individuals, the rest being owned by companies, groups or partnerships. The same study found that portfolio sizes were small: a quarter of all landlords had only one letting throughout Scotland and a half had fewer than 10. Our evidence shows that individual ownership has grown and that portfolio sizes have fallen since 1992/93.

As Table 1 shows, a very large majority of dwellings (84 per cent) in 2008 were owned by private individuals, couples or families, all on an unincorporated basis. Private individuals owned 43 per cent and couples and families owned 41 per cent. Incorporated for-profit bodies, including private and public companies, partnership and property trusts, owned 14 per cent of the stock.

Table 1.

Ownership of dwellings by type of landlord and investment orientation

Category	Proportion of dwellings owned (percentage)
Type of landlord
Individual or couple	84
Company, partnership or property trust	14
Institution	2
All (N = 1463 weighted)	100
Investment orientation of landlord
Business	7
Sideline investor	65
Sideline non-investor	26
Organisation	2
All (N = 1352 weighted)	100

Another way of looking at the types of landlords is to examine whether they operate as full-time or part-time landlords. For individuals, this means that their full-time occupation is being a landlord. For companies or institutions, it means that their main business is letting residential accommodation. Only 8 per cent of dwellings had full-time landlords. Dwellings owned by companies, partnership and property trusts were more likely to have full-time landlords, but only 32 per cent ofthem let accommodation as their main business.

5.2 Investment orientation

We used a fourfold classification to examine landlords’ investment orientation (see Crook et al., 2001; Thomas and Snape, 1995). The four categories were

business landlords: for whom letting accommodation is a full-time occupation or main business;

sideline investor landlords: for whom letting is part time (or not their main business) but who seek investment returns;

sideline non-investor landlords: part-time landlords (or companies for whom letting is not their main business) who do it for non-investment reasons (for example, to house relatives or employees); and

organisational landlords: those institutions holding property for non-investment reasons (such as for housing ministers of a church).

Table 1 shows that business landlords owned 7 per cent of all dwellings. Even fewer, less than 2 per cent, were owned by organisations such as religious institutions or charitable trusts run for not-for-profit purposes (such as housing people in need). Hence, sideline landlords owned the great majority (91 per cent) of privately rented dwellings in Scotland, whether they were individuals, couples, companies, partnerships or property trusts. Most of the dwellings owned by sideline landlords were regarded as investments, but just over quarter (26 per cent) were owned for other reasons, including as a current or future home for the landlord or a relative.

Irrespective of the business orientation of the owners, 72 per cent of addresses were regarded as investments: 23 per cent for capital growth alone, 8 per cent for rental income alone and 41 per cent for a combination of both. Taking business landlords and sideline investors together reveals that 72 per cent of Scotland’s private rented sector was owned for business and investment purposes. This is a significant change from 1992/93 and also compares with recent analyses in England, where there has also been a significant increase in the proportion of the sector regarded for investment purposes (Crook and Kemp, 2010).

5.3 Market Entry and Portfolio Size

Many dwellings were owned by landlords who had first started in the letting business very recently. Only 14 per cent of dwellings had landlords who first started letting before rent deregulation in 1989 and two-thirds were owned by those who first started in the past decade: that is, after 1998 (including 43 per cent in the previous five years). This suggests a large corps of relatively new landlords who, in contrast to more longstanding ones, may not have the risk-spreading advantages of large portfolios or possess significant bodies of collected information or ‘tacit knowledge’ about their market areas.

Despite the investment focus of most landlords, property portfolios are small. The mean portfolio size of all landlords is 12 dwellings but the median is 2 and very few have portfolios in excess of 20 dwellings (Table 2). Not unexpectedly, dwellings owned by company and institutional landlords are more likely to be parts of larger portfolios than dwellings owned by individual landlords. Thus, while the median ownership of individuals and couples was 2, for companies it was 8 and for institutions it was 13 (the means were 5, 29 and 259 respectively). The modal portfolio size for all landlord types is 1. This evidence confirms that portfolio sizes have fallen in recent years in comparison with earlier findings (Kemp and Rhodes, 1994).

Table 2.

Use of managing agents and portfolio size by landlord type

	Type of landlord
	Individual or couple	Company, partnership or property trust	Institution	All
Portfolio size (percentage of dwellings)
1 dwelling	44	16	13	39
2–4 dwellings	36	21	9	34
5–19 dwellings	17	38	39	20
20–99 dwellings	3	20	22	5
100 or more dwellings	<1	5	17	1
N (weighted)	1214	200	123	1437
Use of managing agents (percentage of dwellings)
Managed wholly by landlord	48	63	46	50
Managed partly by agent	18	21	31	18
Managed wholly by agent	35	17	23	32
N (weighted)	1156	192	26	1374

Note: column percentages may not sum to 100 due to rounding.

5.4 Portfolio Management

Half of the sample dwellings were managed solely by their landlords (see Table 2). Agents were involved in the management of the remaining 50 per cent of dwellings, with 32 per cent of all dwellings having agents who undertook all the management tasks on behalf of the landlords and 18 per cent where agents undertook some of the tasks. Dwellings with business landlords were also more likely to be self-managed by the landlord (68 per cent). As Table 3 reveals, landlords with larger portfolios were more likely to manage properties themselves, and the use of agents for all management tasks was most prevalent among landlords of smaller portfolios.

Table 3.

Use of managing agents by portfolio size

Portfolio size (total dwellings in Scotland)	Proportion of dwellings (percentage)			Sample numbers (weighted)
	Managed wholly by landlord	Managed partly by agent	Managed wholly by agent
1	37	18	45	539
2–4	58	16	26	450
5–19	59	21	20	280
20+	67	18	15	84
All	50	18	32	1 353

6. The Management of Market and Business Risk

The evidence presented thus far shows very clearly that by 2008 the private rented sector in Scotland had become increasingly owned by individual landlords whose portfolios were smaller in size than in previous years, although far more were then owned for investment purposes. Hence the market and business risks of being a landlord still fall mainly on the smaller landlords and unincorporated individuals. These findings are important because smaller portfolios will ceteris paribus increase exposure to the ‘normal’ business risks associated with letting. Having 100 properties where 1 per cent of tenants pose problems is a risk that can be managed, but a landlord with only one property faces a much bigger risk if their sole tenant creates a problem (such as rent arrears or anti-social behaviour). As a consequence, it might be reasonably expected that small landlords need to be significantly more careful about the market and business risks that they take.

Handing the effort over to managing agents removes some of the burden of managing business risks, but in return for a fee of usually at least 10 per cent of the gross rent plus expenses. Residential lettings management remains a largely unregulated profession, although there have been voluntary efforts made to professionalise the sector and there is a voluntary system of codes of conduct (Law Commission, 2007). We found that only 15 per cent of dwellings belonged to landlords who were members of a landlords’ or agents’ association or of a relevant professional institute. In line with previous research, we found that, where agents were not used, the most common source of advice to landlords was friends (cited by landlords of 43 per cent of the stock) and also that many landlords found it difficult to find out how the law affected them.

6.1 Distance between Landlord Address and Property

In light of this, the management of the various market and business risks is clearly central to the sector’s viability. What we know about the role of sub-markets, informational economics and landlords’ behaviour suggests a hypothesis that detailed knowledge of the areas or market sectors where property is acquired and managed is a significant risk mitigation strategy used by landlords. Table 4 shows the results of a spatial analysis of landlord and portfolio postcodes. It confirms that landlords generally live near their portfolios and that those managing properties entirely by themselves lived closer than landlords using agents. Even where agents are used, however, landlords still lived near their dwellings. This supports the findings of Rhodes and Bevan’s (2003) study of ‘Buy to Let’ landlords and suggests that the same spatial relations apply within the private rented sector more generally.

Table 4.

Distance between landlord and property

Landlord characteristics	Distance (km)					One-way ANOVA (mean)
	Mean	Maximum	Median	Percentile 25	S.D.
Total dwellings in Scotland
1	66.55	761.67	2.84	0.70	152.10
2–4	32.83	633.88	2.81	0.50	93.14
5–19	22.45	531.76	2.82	0.69	70.79
20–99	16.07	146.17	4.22	1.93	30.00	F = 17.868
100+	16.64	151.35	6.56	2.52	36.48	p = 0.000
Landlord type (a)
Private individual orcouple, etc.	43.80	761.67	2.67	1.01	118.54
Company	12.65	555.71	6.56	1.62	37.02	F = 15.038
Organisation	67.07	151.35	20.71	9.22	68.92	p = 0.000
Landlord type (b)
Business landlord	20.74	531.76	6.56	4.24	50.03
Sideline investor landlord	33.07	648.60	2.52	0.97	98.20
Sideline non-investor landlord	67.13	761.67	2.49	0.00	152.09	F = 14.884
Organisational landlord	20.50	141.05	9.22	9.22	33.42	p = 0.000
Location
Major urban area	35.74	617.35	2.74	2.01	102.96
City	60.56	648.60	2.70	1.10	125.61
Other central belt	24.51	576.62	6.56	1.26	84.85	F = 15.751
Rural	40.02	761.67	3.24	0.27	107.93	p = 0.000
Date landlord first let
Before 1989	31.17	531.76	3.38	0.96	75.94
1989–1995	36.42	540.46	2.67	0.63	86.52
1996–2002	37.69	636.95	3.61	1.02	105.23	F = 11.273
2003–2008	56.21	761.67	3.28	0.82	138.58	p = 0.000
Use of managing agents
Manage it entirely by myself	43.53	934.28	2.01	0.00	150.63
Managing agent does some tasks	78.48	894.25	4.18	0.53	171.67	F = 3.571
Managing agent does all tasks	54.13	897.19	5.68	2.02	156.58	p = 0.028
All	36.19	761.67	3.66	1.12	102.45

Mean distances between property and landlord were heavily skewed by a small number of very distant landlords. Hence, while the mean distance between sample property and landlords resident in the UK was 36.2 km, the median distance was only 3.7 km.

Among landlords of smaller portfolios (<20 dwellings), there was no discernible effect of portfolio size on distance. A quarter of smaller landlords’ properties were situated within 1 km and half were within 3 km. However, larger landlords appear to have had slightly more dispersed stockholdings. This is an obvious consequence of their size and may also reflect their desire to own ‘pepper-potted’ property rather than geographically concentrated blocks of properties. However, even the very largest landlords (of ≥100 dwellings) had a quarter of their properties within 2.5 km and half within 6.6 km. Despite having greater median distances, larger landlords had smaller mean distances—this suggests that, generally, larger landlords consolidate their stock spatially while smaller landlords, especially those with only one dwelling, can live a significant distance from their properties. This may be a reflection of particular circumstances, such as parents owning a dwelling to house a child at university, or temporarily letting out their previous home whilst moving away for work.

A similar pattern emerged with landlord types. Private individuals and couples were on average 43.8 km away from their property, yet a quarter were within 1 km and half within 2.7 km. Companies, on the other hand, were on average much closer to their properties (12.6 km), although a quarter of their properties were within 2.5 km and half were within 6.6 km. Organisational landlords were much more distant from their stock than either individuals or companies. On average, landlords with investment or business motivations tended to be much closer to their stock than ‘sideline non-investor’ landlords. Meanwhile, established landlords were on average closer to their stock than more recent entrants to the sector. Properties in rural areas were not on average any further from their landlords than in other types of area despite lower settlement densities.

Finally, those landlords that use managing agents for some or all tasks tended to be further away from their properties. The median distance between landlord and property for those landlords not using managing agents at all was 2 km. When a managing agent was employed for some tasks, the median distance rose to just over 4 km and when a managing agent did all the tasks, the median distance was just under 6 km.

6.2 Living near Portfolios

The importance of living near to one’s portfolio was discussed in greater detail in a series of focus groups with landlords. Respondents revealed how important knowing an area and living close to it was to the way they managed their portfolios and the risks they faced. This was as true for the acquisition of properties and the mitigation of market risks as it was for managing day-to-day business risks.

Respondents generally indicated that a key way of managing the market risks associated with acquiring properties was to know personally the sub-market where property was acquired. Landlords (of all types and portfolio sizes) said they needed to live near their properties to do this. This helped them to manage market risk by using local knowledge to spot good-value acquisitions and to observe closely how the property market changed. It also helped them to manage business risks by personally addressing property management issues (often even when a managing agent was employed).

Managing agents were often criticised for poor performance. Most landlords who used them (even if they had a good experience) checked their own properties personally and kept in touch with tenants. As one landlord put it

There is a lack of professionalism, some don’t know how to manage and even the vetting of tenants has not been successful (Stirling landlord).

A number of landlords had dispensed with managing agents and took the management of their stock back into their own hands. Even where agents’ performance was good, landlords found that their fees (alongside repairs and insurance costs) so reduced their gross income returns from rent that they had to rely on capital appreciation for their return. Ensuring a reasonable return on investment was one reason why so many landlords undertook their own property management. Where agents were used it was often simply to find tenants, with the landlord subsequently managing the property.

Many tenants came by ‘word of mouth’ and not through advertising. Although landlords regularly used formal vetting procedures (such as credit reference and other agencies), they rarely found them reliable guides to the ability of tenants to pay rent or look after flats and houses. Landlords therefore placed a heavy reliance on meeting prospective tenants themselves—and knowing the local area. Almost everyone did a pre-letting inventory (often using digital cameras).

Once again, these mitigation practices for business risks generally required landlords to live near their portfolios. One landlord adopted the specific criterion that all acquisitions were within “one hour’s drive of where I live” (Aberdeen landlord). The concept of being able easily to ‘keep an eye’ on property was a recurrent theme. This applied even when agents were used

It’s important to keep an eye on it especially in between tenancies; even if you have an agent … you can’t get in when it’s let (Edinburgh landlord).

As is clear, a proximate portfolio imparts distinct advantages when it comes to the day-to-day management of property and mitigating business risks. Our respondents made it clear that proximity was helpful in mitigating market risks, too. Landlords felt that they knew their own local areas well and hence how they performed in housing market terms

It’s important to know the local area … they are cycling distance from my home and I understand the market and know the area (Edinburgh landlord).

The ability to recognise the distinctive micro geography of housing markets was a key characteristic among respondents, even among large landlords

What’s important to know is to have local knowledge because [the] difference between one street and [the] adjoining street in terms of value and potential for renting are [sic] enormous (Glasgow landlord).

The landlords who did not live near their properties, on the other hand, were relatively rare. Those we spoke to owning properties well beyond their immediate neighbourhood universally used managing agents to deal with those properties—but were also careful to buy properties in those more distant locations that “run themselves” (Fife landlord). In many cases, these were landlords who had acquired property to house immediate family (often children at university) or to let their own home temporarily whilst they had relocated to other cities for work.

7. Conclusions and Policy Implications

The ambitions of governments to stimulate the emergence of larger and corporate landlords, able to use their size to achieve economies of scale and to manage risks through market diversification, have so far not been realised. This raises the question about the risks to investment that landlords face and the strategies they might adopt to minimise them. Private rental markets are dominated by small landlords. There are significant implications for wider housing policy if those landlords are unable to expand in a way that involves acceptable levels of risk.

Our results show that the general tendency is for landlords to restrict their operations to areas they know well, close to where they live or work and where they know something of the property market. They are often in touch with informal networks in neighbourhoods that give them signals about rents being charged by other landlords and about properties that might be coming onto the market in the near future. In many cases, landlords know their tenants’ neighbours and these are useful sources of information about their own tenants’ behaviour. Living nearby enables landlords to keep an eye on their properties in relation to repairs issues or to detect tenancy abandonments. While the use of managing agents allows landlords to be more distant from their properties, the quality of agency services varies and in many cases they are unused or seen as unaffordable.

These findings are consistent with a conceptualisation of the housing market that emphasises the importance of distinctions between areas and sub-markets even within the same local housing market, and the importance of local and sub-market-specific information flows. Such information flows suffuse landlords’ behaviour and support their decision-making. These behaviours are necessary to mitigate risk—both related to capital investment (market risk) and the day-to-day management of a property portfolio (business risk). The market risks in particular arise because of the segmented nature of local housing markets, making an understanding of price formation critical. In turn, the localised investment behaviour of landlords reduces the level of spatial arbitrage that occurs in the market, which is likely to contribute to the persistence of sub-markets. More fundamentally, it appears that developing good local knowledge is a key strategy for running a small residential lettings business.

One of the reasons for encouraging larger, corporately financed landlords is that they are assumed to be better placed to manage uncertainty through risk diversification (by having different types of property in different locations) and that they will be in a better position to acquire information on the basis of their size and may have an ability to provide agency services in-house. That this development has not yet happened means that the risks (and rewards) of providing private residential lettings fall on small landlords whose risk mitigation strategies mean that they operate within a very restricted geographical area. In these areas, there is a limit to the number of properties any one landlord will want to own because of the risks of market saturation. Yet landlords may also be reluctant to ‘stray’ outside the narrow confines of well-known neighbourhoods of which they have detailed personal knowledge. This risk management approach thus limits the capacity for the existing body of landlords to grow the sector from within. Expansion of the sector would therefore require new entrants.

The policy implications of our findings are perhaps three-fold. First, it appears that securing greater professionalism among managing agents would be important, so that landlords can be more confident about accessing good asset and housing management services and, in particular, making better-informed judgements about holding part of their portfolio in areas not known to them personally. Amongst other measures, this suggests some statutory regulation of these services. Secondly, improving information flows in the sector will be important in tackling asymmetries between actors and between areas in the market. Improved peer group benchmarking services, along the lines of those used by commercial property landlords, could take the form of rental or yields indices and other micro-economic information made available both to landlords and (prospective) tenants to diffuse knowledge about otherwise remote sub-markets. We found no evidence that landlords systematically used commercial data services and it is clear that (in the UK at least) such services are neither targeted at, nor priced for, small landlords. However, the opening up of government data on rents used to administer the Local Housing Allowance for private tenants has helped to provide information about that part of the market. Thirdly, our findings reinforce the importance of addressing barriers to getting larger corporate landlords financed with institutional funding. Such landlords may be better able to address the risk constraints that small landlords face (Crook and Kemp, 2002). Together, these three approaches would diminish the importance of having a geographically proximate stock portfolio, thus removing some of the barriers to the growth of private rented housing in those markets where the objective is to seek its expansion.

Our survey findings suggest that the geographical relations between landlords and portfolios are of importance. Qualitative insights from focus group participants confirmed this. However, we would suggest that there are two key areas for further research. First, our findings suggest the need for a more formal treatment of the role of landlord–portfolio relations in sub-market models. This could include analyses of the neighbourhood types favoured by landlords with different characteristics. Secondly, closer links between our findings and studies of demand-side behaviours in local markets would be fruitful. Both questions would benefit from explicit recognition of the role of geographical proximity as an informational strategy among market actors.

Footnotes

Appendix

Acknowledgements

The authors wish to thank four anonymous referees for their helpful comments on the paper. The views expressed by the authors are their own and do not necessarily reflect the views of the Government. As usual, responsibility for any errors and omissions rests with the authors.

Funding Statement

The authors wish to acknowledge the support of the Scottish Government who funded some of the research reported in this article.

References

Allen

M. T.

Springer

T. M.

Waller

N. G.

(1995) Implicit pricing across residential rental submarkets, Journal of Real Estate Finance and Economics, 11, pp. 137–151.

Best

Kemp

Coleman

(1992) The Future of Private Renting: Consensus and Action. York: Joseph Rowntree Foundation.

Bevan

Kemp

P. A.

Rhodes

(1995) Private landlords and housing benefit. Centre for Housing Policy, University of York.

Crook

A. D. H.

Kemp

P. A.

(1996) The revival of private rented housing in Britain, Housing Studies, 11(1), pp. 51–68.

Crook

A. D. H.

Kemp

P. A.

(2002) Housing investment trusts: a new structure of rental housing provision?, Housing Studies, 17(5), pp. 741–754.

Crook

A. D. H.

Kemp

P. A.

(2010) Transforming Private Landlords. Oxford: Wiley Blackwell.

Crook

A. D. H.

Kemp

P. A.

with Barnes, Y. and Ward, J. (2002) Investment returns in the private rented sector. British Property Foundation, London.

Crook

A. D. H.

Kemp

P. A.

Ferrari

E. T.

(2009) Views and Experiences of Landlords in the Private Rented Sector in Scotland: Scottish Government Review of the Private Rented Sector, Volume 3. Edinburgh: Scottish Government.

Crook

A. D. H.

Henneberry

J. M.

Hughes

J. E. T.

Kemp

P. A.

(2001) Repair and maintenance by private landlords. Department of the Environment, Transport and the Regions, London.

10.

Dixon

Leach

(1978) Sampling methods for geographical research. CATMOG No. 17, Study Group in Quantitative Methods of the Institute of British Geographers Geo Abstracts, University of East Anglia.

11.

Galster

(1996) William Grigsby and the analysis of housing sub-markets and filtering, Urban Studies, 33(10), pp. 1997–1805.

12.

Garmaise

M. J.

Moskowitz

T. J.

(2004) Confronting information asymmetries: evidence from real estate markets, The Review of Financial Studies, 17(2), pp. 405–437.

13.

Gibb

(1994) Before and after deregulation: market renting in Glasgow and Edinburgh, Urban Studies, 31(9), pp. 1481–1495.

14.

Gibb

Nygaard

(2004) Buy to let and the Glasgow housing market. Glasgow University (mimeograph).

15.

Grigsby

W. G.

(1963) Housing Markets and Public Policy. Philadelphia, PA: University of Pennsylvania Press.

16.

Hansard (2010). Answer by the Minister for Housing, Rt Hon. Grant Shapps MP HCDeb 10 June 2010 c446.

17.

Hickman

Sprigings

Cole

(2008) The private rented sector in West Yorkshire. Final report, Centre for Regional Economic and Social Research, Sheffield Hallam University.

18.

HM Treasury (2010) Government response to the consultation on investment in the private rented sector. HM Treasury, London.

19.

Jones

MacDonald

(2004) Sustainable urban form and real estate markets. Paper presented to the European Real Estate Conference, Milan, June.

20.

Jones

Leishman

Watkins

(2002) Examining the persistence of submarket price differences. Paper presented to the Pacific Rim Real Estate Society Conference, Christchurch, January.

21.

Jones

Leishman

Watkins

(2003) Structural change in a local urban housing market, Environment and Planning A, 35(7), pp. 1315–1326.

22.

Jones

Leishman

Watkins

(2004) Migration linkages between urban housing markets: theory and evidence, Housing Studies, 19(4), pp. 269–283.

23.

Kemp

P. A.

(2004) Private renting in transition. Chartered Institute of Housing, Coventry.

24.

Kemp

P. A.

Rhodes

(1994) Private landlords in Scotland. Scottish Homes, Edinburgh.

25.

Kemp

P. A.

Rhodes

(1997) The motivations and attitudes to letting of private landlords in Scotland, Journal of Property Research, 14(2), pp. 117–132.

26.

Law Commission (2007) Encouraging responsible renting. The Law Commission, London.

27.

Leyshon

French

(2009) ‘We all live in a Robbie Fowler house’: the geographies of the buy to let market in the UK, The British Journal of Politics and International Relations, 11(3), pp. 438–460.

28.

Lowe

(2007) The new private rented sector: regulation in a deregulated market, in: Hughes

Lowe

(Eds) The Private Rented Housing Market, pp. 1–14. Aldershot: Ashgate.

29.

Maclennan

(1982) Housing Economics. Harlow: Longman.

30.

Maclennan

(1996) Economic perspectives on the structure of local housing markets, Housing Studies, 11(3), pp. 387–406.

31.

Meen

(2001) Modelling Spatial Housing Markets. London: Kluwer.

32.

Pawson

(2007) Local authority homelessness prevention in England: empowering consumers or denying rights?, Housing Studies, 22(6), pp. 867–883.

33.

Priemus

Maclennan

(1998) The different faces of private rented housing, Netherlands Journal of Housing and the Built Environment, 13(3), pp. 99–104.

34.

Read

(1997) Vacancies and rent dispersion in a stochastic search model with generalized tenant demand, Journal of Real Estate Finance and Economics, 15(3), pp. 223–237.

35.

Rhodes

Bevan

(2003) Private landlords and buy to let. Centre for Housing Policy, University of York.

36.

Rhodes

Kemp

P. A.

(2002) Rents and returns in the residential lettings market, in: Lowe

Hughes

(Eds) The Private Rented Sector in a New Century, pp. 43–64. Bristol: The Policy Press.

37.

Rugg

Rhodes

(2008) The private rented sector: its contribution and potential. Centre for Housing Policy, University of York.

38.

Rutherford

R. C.

Springer

T. M.

Yavas

(2007) Evidence of information asymmetries in the market for residential condominiums, Journal of Real Estate Finance Economics, 35, pp. 23–38.

39.

Scottish Government (2007) Firm foundations: the future of housing in Scotland: a discussion document. Scottish Government, Edinburgh.

40.

Scottish Government (2009) Scottish government review of the private rented sector. Scottish Government, Edinburgh.

41.

Stigler

G. J.

(1961) The economics of information, The Journal of Political Economy, 69(3), pp. 213–225.

42.

Thomas

Snape

(1995) In from the cold? Working with the private landlord. Department of the Environment, London.

43.

Wallace

(2008) Knowing the market? Understanding and performing York’s housing, Housing Studies, 23(2), pp. 253–270.

44.

Watkins

C. A.

(2008) Microeconomic perspectives on the structure and operation of local housing markets, Housing Studies, 23(2), pp. 163–177.

45.

Wheaton

W. C.

(1990) Vacancy, search, and prices in a housing market matching model, Journal of Political Economy, 98(6), pp. 1270–1292.

46.

Yavas

(2001) Resolution of demand uncertainty and competitive rent adjustment, Journal of Housing Economics, 10, pp. 472–481.