Migration within England and Wales and the Housing Market

Migration within England and Wales and the Housing Market....... Economic conditions exert a strong influence on regional migration. On the one hand, strong labour market conditions, as exemplified by low unemployment rates and high earnings, draw migrants into regions. On the other hand, strong housing market conditions can prevent movement since commuting may often be an alternative to migration. This can be thought of as giving rise to a migration equilibrium where high house prices choke off migration caused by strong labour market conditions. Expected capital gains in housing, however, can offset high levels of house prices, an effect ignored in previous literature. Migration can also be influenced more directly by the availability of housing relative to population without this being mediated through prices. This paper, by Gavin Cameron, John Muellbauer and Anthony Murphy, presents evidence on inter-regional net and gross migration between the regions of England and Wales that is broadly in accord with these expectations.... I. Introduction Migration between regions plays a central role in the workings of regional housing and labour markets. As such, it has often been said that the relatively low level of labour mobility in the UK is intimately connected with the poor performance of some regional labour markets and the relatively unusual structure of housing tenure in the UK, with its small private rental sector and declining social rented sector (see Hughes and McCormick (1987, 2000), and McCormick (1997)). The high degree of persistence in regional unemployment rates compared with the US is perhaps one symptom of this inefficiency (Cameron and Muellbauer (2001)). Bover et al. (1989) and Muellbauer and Murphy (1991) suggested a number of ways that a large owner-occupied sector could exacerbate regional mismatch over and above the fact that owners face higher transactions costs in moving. A body of econometric evidence has now built up to suggest that high relative earnings and employment opportunities encourage migration to a region, while high relative house prices discourage it. The most obvious mechanism for this negative effect of house prices on migration arises through cost of living differentials between regions. If earnings are not deflated by a regional cost of living index which appropriately incorporates housing costs, relative house prices will, in part, measure this omitted cost of living effect. However, there are likely to be further effects connected with constraints on credit availability and the risks associated with a high level of indebtedness both relative to income and to housing equity. In addition, Muellbauer and Murphy (1997) show that the portfolio motive for house purchase is very important in Britain. Given this importance, the demand for living space by employees can sometimes be crowded out by higher portfolio demand. In this paper we show that the expected relative appreciation of house prices has an important effect on migration. This helps explain the peak of net out-migration from the South East in 1987-89 when South East labour demand was so buoyant: at this time high relative house prices coincided with high appreciation expected in other regions, both factors encouraging net outmigration. This is an important example of regional labour mismatch resulting both in efficiency losses and inflationary pressure. Of course, migration on its own is only part of the wider interaction of labour and housing markets, in which wages, unemployment and house prices, as well as migration, are jointly determined. This paper models regional migration choices as a system of eight regional equations for England and Wales (that is, excluding Scotland) with crossequation restrictions. Data are taken from the National Health Service Central Register for 1975 to 2003 and we examine both net and gross migration rates. Gavin Cameron, Department of Economics, University of Oxford, John Muellbauer, Nuffield College, University of Oxford, Anthony Murphy, University College, Dublin. Acknowledgements: This paper builds on a wider project funded by the ODPM on Affordability Targets: Implications for Housing Supply (www.odpm.gov.uk). We would like to thank Paul Chamberlain, David Hendry, Michael Kell, Geoff Meen, and participants at seminars in Oxford and Reading, for helpful comments. All errors and omissions remain our own. July 2005 9 Migration within England & Wales

II. Some views from the literature In this section we provide a brief overview of some recent research on migration within Britain. Three main themes have emerged in this recent research, which we cover here in turn: the labour market, the housing market, and international migration. Hughes and McCormick (2000) use data on individuals from the Survey of English Housing and the Labour Force Survey to examine a bundle of issues relating to both labour and housing markets. Their data suggest that overall migration rates were relatively high in the 1990s, with renters and less-skilled workers having become increasingly mobile since the 1970s and owners slightly less mobile. Their econometric results, based on individuals, find that relative, regional wages, but not relative regional unemployment rates, significantly influence migration between regions. At an individual level, being unemployed tends to make a person more likely to migrate out of a region. They extend their analysis to low-income households (household income less than 10,000 per annum) and show that, all else equal, private tenants are about fifty times more likely to migrate than low-income local authority tenants. Lowincome tenant households are also more likely to migrate from a region which has a relatively high price of owner-occupied housing. Turning to direct housing market influences, they find that, even after controlling for other influences such as age, socio-economic group and household status, the differences in mobility by tenure remain. Local authority tenants have the lowest migration rates, followed by owner-occupiers. Tenants of social landlords and, particularly, private tenants are found to be the most geographically mobile. Interestingly, controlling for other influences, local authority tenants have become more mobile over time and their migration rate is now only slightly lower than that of owner-occupiers. In general, they also find that there is a higher migration rate out of regions with higher house prices, controlling for other influences such as regional wage rates. The effects are greatest for people approaching retirement, but in more recent years, the relationship seems to have weakened. Since international migration is a politically divisive topic, it is natural to expect that even the economics literature will be divided to some extent (compare Borjas,(2003), with Dustmann et al.(2003), for example). This is certainly the case, with a fair variety of different conclusions being drawn by different studies. It is worth stating baldly, however, that economic theory provides little support for the idea that significant levels of international migration will be without negative effects on the wages and employment of competing workers. In a major review of the UK labour market consequences of international migration, Dustmann et al. (2003) find that there was a positive but mostly insignificant effect on unemployment across the regions of Britain, and a small positive effect on wages. Dustmann et al. suggest that the positive effect may arise from the fact that migrants may have skills that are complementary to domestic workers. On the other hand, it seems just as plausible to think that it may be due to problems inherent in the nature of the estimation of the local effects of migration when there are many adjustment mechanisms that might lead to the effect being spread more widely than just locally, especially because there is a general equilibrium response to migration that leads to onward migration by locals to other regions, thereby spreading the effect across the whole economy. 1 As noted by Hatton and Tani (2003), migration into the UK has increased considerably in the past decade or so, with around 110,000 migrants arriving each year in the second half of the 1990s, compared with around 37,000 in the second half of the 1970s. Hatton and Tani (2003) take this as a starting point for an analysis of the effect of international migration on net inter-regional inmigration rates in the Standard Statistical Regions of Great Britain from 1982 to 2000. Although their estimates are not very precise, they typically find that the arrival of an additional 100 immigrants to a region results in the displacement 1 Most of the international migration literature looks at the effect of migration on local labour markets for unskilled workers. This is problematic for a number of reasons. First, because migrants may be attracted to cities with thriving economies, there may be a spurious correlation between wages and migration that is hard to completely control for econometrically. Second, because migrant education, skill and experience levels are rather diverse, their effects on the labour market may not be easily captured by the simple dichotomy of skilled and unskilled workers. Third, because there may be general equilibrium responses to migration, including the upgrading of skills and capital accumulation by locals, and also out-migration to other regions or falling participation, which may attenuate the effect in other regions. Borjas (2003) makes a valiant attempt to control for these influences and concludes that for the USA between 1980 and 2000, a 10 per cent increase in migrant supply reduces wages by 3 to 4 per cent. Migration within England & Wales 10 July 2005

to another region of between one-third and twothirds of that number. This therefore provides a mechanism by which the labour market effect can be attenuated in the region in which international migrants arrive, - although given the notorious lack of supply response in the British housing market (see Barker, 2004), it is perhaps not surprising to find a displacement effect of this magnitude. III. Data Issues Annual (indeed quarterly) information on regional migration flows is available back to 1976 from the Office of Population Census and Surveys (OPCS). These data are based on National Health Service records of patients registering with local doctors and are adjusted for small lags between moving and registering, (see Hornsey (1993) for discussion of the data). Since 1982, an age breakdown of migrants has been published. This suggests that generally under 12% of migrants are over retirement age (65 for men, 60 for women), while under 20% are aged 15 years or less. Thus, migrants consist largely of working age adults and their dependents. Table 1 shows the total number of people moving between various Government Office Regions of the UK in 2002. Since the definition of regions changed from Standard Statistical Regions to Government Office Regions in early 1990s, in our later analysis we merge North East and North West into total North (NT), and East and Rest of South East into total South (ST). Table 1 Interregional movements, 1,2 2002 Area of origin Thousands United Kingdom England North East North West Yorkshire and the Humber East Midlands West Midlands East London South East West Wales Scotland Northern Ireland United Kingdom 3-119 41 108 95 97 103 130 262 220 111 50 48 11 Area of destination England 101-36 87 86 90 89 120 247 199 94 48 45 9 North East 43 37-6 9 3 3 3 5 5 2 1 4 1 North West 109 91 6-18 10 13 8 14 14 8 9 7 2 Yorkshire and the Humber 100 92 9 18-17 8 9 11 13 6 3 5 1 East Midlands 120 113 3 10 18-17 20 15 22 8 3 3 1 West Midlands 99 87 3 12 8 14-8 13 16 13 8 3 1 East 150 142 3 7 7 13 7-66 28 9 3 4 1 London 155 142 5 12 10 10 11 28-51 15 5 7 1 South East 229 213 4 12 9 14 14 29 98-33 8 7 1 West 146 132 2 9 6 9 16 14 25 50-9 4 1 Wales 64 62 1 11 3 3 10 4 6 11 12-2 - Scotland 53 49 4 8 5 3 3 5 7 9 4 2-2 Northern Ireland 11 8-1 1 1 1 1 2 1 1-2 - 1 Based on patients re-registering with NHS doctors in other parts of the United Kingdom. Figures have been adjusted for minor changes caused by database realignment during Health Authority organisation. 2 At the time this table was being prepared for publication, research was ongoing into whether there was a need to revise internal migration estimates in the light of the results of the 2001 Census. 3 Total number of people moving from other parts of the United Kingdom Source: Regional Trends, Office of National Statistics, 2004. National Health Service Central Register; General Register Office for Scotland; Northern Ireland Statistics and Research Agency. July 2005 11 Migration within England & Wales

IV. Methodology This paper reports the results of part of a larger project that attempts to model the important feedbacks between migration, household formation, house prices, tenure choice, earnings, employment and unemployment. This systems approach is necessary when one considers how to answer a question such as whether an increase in South East housing supply will have an effect on house prices since, as well as the effects within the South East housing market (on tenure choice, household formation and the excess demand for housing directly), it will also have an impact on the South East labour market, and also on migration into and out of the region, especially on migration in and out of Greater London. V. Outline structure of the model The theoretical basis of the model is fairly simple. Consider a utility maximising household making a location choice. High relative earnings and employment, along with other local amenitie,s will tend to attract households, while high relative house prices will tend to deter them both because of cost-of living effects and because of creditconstraints operating through the mortgage lending system. In addition, a lower local population density (per unit of housing) is likely to attract migrants, while a recent history of negative returns in housing will deter a risk-averse household (given the short-to medium-run persistence of housing returns). This can seen as giving a kind of amenity-cost trade-off that pitches the labour market amenity against the housing market cost, with other regional characteristics captured by economic variables described below, or fixed effects. 2 However, there is one important, and neglected, positive aspect of the housing market, - namely, that the expectation of house price appreciation may be able to overcome high house prices (see Kiel (1994) and Cameron and Muellbauer (1998)). We return to this issue in the empirical results below. Since the transactions cost of moving (especially before the arrival of the internet) increase with the distance to be moved, one would also expect contiguity to be important in regional location decisions. Consequently, relative house prices in 2 This is in the spirit of the migration equilibrium approach of Harris and Todaro (1970,) who argued that the expected wage in urban areas (of developing countries), adjusted for the unemployment rate, should be equal to the marginal product of an agricultural worker. contiguous regions should matter more. However, in the labour market, the commuting-migration trade-off can weaken or reverse the role of near versus far alternatives (Gordon (1975, 1990), Gordon & Molho (1998), Molho (1982), Jackman and Savouri (1992a and 1992b)). The analysis below also explores the related issue of connectivity. This is the idea that, for example, the South West is rather less integrated into the national economy given its spatial separation. So, one might expect regional interactions to be affected by how connected a region is with its neighbours. It turns out that the basic results are not much affected by introducing this effect. In the following analysis, we have two ways of measuring working-age migration and, hence, two dependent variables - namely net and gross inmigration into a region. For modelling purposes, these two variables are both scaled by gross migration between all British regions and by the regional working age population. For example, inmiga r = (in r / wapop r ) / [(migsum /migsum 1990 ) / (wapop gb / wapop gb,1990 )] where inmiga is in-migration (respectively net or gross), wapop is working-age population and migsum is the sum of all migration within Britain. Essentially this provides a measure of in-migration per person in the working-age population, relative to Britain. Turning now to the formulation of the independent variables, we have measures of relative log fulltime earnings and relative log house prices which capture an effect relative to Great Britain and a separate effect relative to contiguous regions (weighted by working age populations): (1-λ 1 )*(lfte r -lfte GB ) + λ 1 *(lfte r -contig lfte r ) Call this F 1 (lfte r ). (1-λ 2 )*(lhp r -lhp GB ) +λ 2 *(lhp r -contig lhp r ) Call this F 2 (lhp r ). We expect λ 2 to be positive and λ 2 to exceed λ 1. λ 1 could be negative. As mentioned above, our analysis also examines an issue related to contiguity, that of connectedness. Some GORs are entirely surrounded by other regions, - for example, the West Midlands and GL, - whereas others, - such as the South West, - share only a relatively small border with other regions. Migration within England & Wales 12 July 2005

Therefore, an alternative to a simple measure of contiguity is to measure connectivity by the ratio of connected boundary to total boundary. For example, the total South region has approximately half of its boundary with other regions, which is scaled up to 0.6 due to the high density of the boundary with Greater London. Overall, the following connectivity weights are adopted, by which the λ s are scaled: North (NT) 0.56 Yorkshire and Humberside (YH) 0.73 West Midlands (WM) 1.00 East Midlands (EM) 0.85 South (ST) 0.60 Greater London (GL) 1.00 South West (SW) 0.33 Wales (WW) 0.36 The gross and net migration equations simplify to partial adjustment equations with a variable speed of adjustment. The speed varies with rate of residential property transactions (ptran). Thus, mig r = α 0r + α(ptran).(x r β mig r,-1 ) where α(ptran) = α 1.(1 + α 2.ln(ptran)) and x r is a set of explanatory variables. We include regionspecific double fixed effects and time trends (see below). We also allow some parameter heterogeneity in the Greater London and South equations. The typical regional in-migration equation is as follows: inmig r = α 0r + α 1.(1 + α 2.lptran).[β 0r inmig r,-1 + β 1.F 2 (rlwapop r,-1 - rlhs r,-1 ) + β 2.F 1 ( rur r ) + β 3.F 1 (rur r,-1 ) + φ.β 4.F 1 (rlfte r ) + β 4.F 2 (MA 2 rlhp r ) + β 5.F 2 (rrhneg r ) + β 6.F 2 ( rlhp1f r ) + β 7r.(year - 1990)] where the subscript r denotes region r, denotes a first difference and the subscript 1 a lag etc. Consider the individual terms in this equation. The first term α 0r is a region fixed effect. The second term α 1.(1 + α 2.lptran) is the variable speed of adjustment term. The coefficient on the speed of adjustment varies with lptran, the log rate of property transactions relative to 1990 (scaled by owner-occupied housing in Great Britain). Lptran is zero in 1990, so α 1 is the speed of adjustment in that year. The first term in square brackets β 0r is a second region fixed effect. This is needed since the means of some of the regressors in the expression in the square bracket could be far from zero. The lagged level of migration inmig r,-1 is an error correction term. The term β 1.F 2 (rlwapop r,-1 - rlhs r,-1 ) refers to the lagged log working age population to housing stock ratio. The more houses there are relative to population, the higher the inflow, other things being equal. The two terms β 2.F 1 ( rur r ) + β 3.F 1 (rur r,-1 ) are relative unemployment rate effects in change and level form. In the inflow equation, the change in unemployment matters more outside London; in London, both matter. Unemployment rates seem to capture labour market effects better than employment rates. The next two terms are relative earnings and house price effects. The term φ.β 4.F 1 (rlfte r ) is the relative log earnings effect. High relative earnings raise inflows, ceteris paribus. β 4.F 2 (MA 2 rlhp r ) is a relative house price effect. High relative house prices reduce inflows, other things being equal. The estimation results suggest that the dynamics for this effect may be represented by a two-year moving average, consistent with decision lags. Thus, φ measures the relative weight on earnings compared to house prices. Downside house price risk is captured by the β 5.F 2 (rrhneg r ) term. Note that rrhneg r is either zero or negative. 3 A negative rate of return on housing has a negative effect on in migration. The β 6.F 2 ( rlhp1f r ) term captures the effects on inflows of expected appreciation in relative house prices next year. Other things being equal, high expected capital gains raise inflows. Finally the β 7r.(year - 1990) term is a region specific trend. The Greater London equation has exactly the same structure except that the β 1 to β 4 parameters are allowed to take on values different from those in other regions. In the South equation, we set β 1 to be the same as in the Greater London equation, β 2 and β 4 take on different values from other regions, and we can accept the restriction that β 2, the unemployment rate level effect, is zero. The out-migration equations have almost the same structure with reverse signs for the economic effects. As noted above, our modelling focussed on a system of eight regional equations for the Government Office Regions of England and Wales, 3 rrh is defined as the rate of capital appreciation plus 3 percentage points for imputed rent minus the mortgage interest rate, so rrhneg=rrh if rrh is negative, or otherwise zero. July 2005 13 Migration within England & Wales

imposing cross-equation restrictions in a two-step manner. First, the Seemingly Unrelated Regressors estimator was used to form an estimate of the covariance matrix of the regression. Second, generalised least squares with robust standard errors was used with the fixed covariance matrix from step 1. Key results are summarized in table 2. To get a feel for the magnitude of the most important housing related effects, we show graphs for three regions. These plot the scaled net migration rate against the composite of the fitted relative earnings and the three types of relative house price effects, and against the fitted effect of houses per head. Figure 1 shows that for the North, much of the variation in the net migration rate is explained by the relative earnings and house price effects, while variations in the log of houses per head only have a relatively small effect. Figure 2, for Greater London, shows exactly the opposite: variations in the number of houses per head, which in recent years has been strongly influenced by international migration, have visibly had a major influence on regional net migration rates. The influence of relative earnings and house prices on net migration is far less obvious, though it helps account for some of the smaller fluctuations. Given space constraints, we do not show the influence of relative unemployment rates, but in London s case, these explain much of the remaining variation in net migration. Figure 3 for the South East shows an intermediate case, where both relative earnings and house prices, and the number of houses per head, play a clearly visible role in explaining variations in net migration rates. Table 2 - Key Parameter Estimates for Regions other than Greater London and the South Net In-Migration Equation α 1 speed of adjustment (in 1990) 0.63 (0.05) β 1 lagged relative log working age pop to -0.027 housing stock ratio (0.016) β 2 change in relative unemployment rate -0.050 (0.011) β 3 lagged relative unemployment rate -0.028 (0.006) β 4 log relative house prices -0.485 (0.174) β 5 relative house price downside risk 1.172 (0.285) β 6 relative expected change in house prices 1.029 one year ahead (0.180) Parameter Estimates Gross In- Migration Equation 0.82 (0.05) -0.012 (0.008) -0.036 (0.010) -0.002 (0.004) -0.435 (0.128) 0.106 (0.278) 0.310 (0.108) Gross Out- Migration Equation 0.58 (0.0) 0.013 (0.005) 0.017 (0.010) 0.025 (0.004) 0.167 (0.082) -0.286 (0.152) -0.381 (0.140) Notes: Standard errors are shown in parentheses. The β 2 to β 6 parameter estimates have all been multiplied by 100. The following admissible restrictions have been imposed - the variable speed parameter α 2 = 0.6, the loadings on contiguous regions λ 1 = -0.6 and λ 2 = 0.4 and the weight on log earnings relative to log house prices ϕ = -2. The estimated β 4 relative house price parameters in Greater London and the South are larger, especially in the out-migration equations. For Greater London, β 1, the population to housing stock ratio effect, is almost three times as large as for other regions. Migration within England & Wales 14 July 2005

0.002 0.001 0.000-0.001 Figure 1 North Region Net Migration of Working Age Population/ Housing stock House prices & earnings 0.002 0.001 0.000-0.001 Figure 4 shows the gross migration between all UK regions and lptran, the log of housing transactions in England and Wales scaled by owner-occupied housing. Recall that we divide the regional migration rates by overall UK migration, and that the speed of adjustment varies with the transactions rate. The decline of lptran implies that the speed of adjustment of migration to housing and labour market effects has fallen since the late 1980s. Figure 4-0.002-0.003 Net migration (scaled) -0.002-0.003-0.004-0.004 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 0.0220 0.0210 Regional Migration & Property Transactions 0.5 0.4 0.3 Figure 2 0.0025 0.0000-0.0025 Greater London Net Migration of Working Age House prices & earnings 0.0025 0.0000-0.0025 0.0200 0.0190 0.0180 Migration rate (lhs axis) Property transactions lptran (rhs axis) 0.0170-0.3 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 0.2 0.1 0-0.1-0.2-0.0050-0.0075-0.0100 Population/ Housing stock -0.0125-0.0125 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 0.006 0.004 0.002 0.000-0.002-0.004 Figure 3 Net migration (scaled) South Net Migration of Working Age Population/ Housing stock House prices & earnings Net migration (scaled) -0.0050-0.0075-0.0100 0.006 0.004 0.002 0.000-0.002-0.004-0.006-0.006 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 Finally, some of the limitations of this kind of exercise must be pointed out. Historical annual data for standard regional classifications on the allocation of rate support grants, infrastructure investment, expenditure under urban renewal schemes, spending by regional development agencies and EU structure fund spending, have not, to our knowledge, been assembled in a systematic form. Such spending could well have implications for regional migration not mediated through unemployment rates, earnings and house prices. Micro-evidence suggests a link between education and mobility. While we have some data on the age composition of migrants, and data on the age composition of regional populations, we do not have historically comparable data on the educational qualifications of either. Further, the composition of housing is not controlled for when we investigate the direct effect of housing availability on migration. Regional fixed effects and trends will not be ideal proxies for such omitted effects. The presence of illegal immigrants and other measurement errors are another limitation. July 2005 15 Migration within England & Wales

VI. Conclusions The models discussed in this paper attempt to capture the effects of labour and housing market features on internal migration within the regions of England and Wales. The paper improves on previous work by Cameron and Muellbauer (1998) in several respects. These are the inclusion of Greater London as a separate region; the modelling of gross as well as net migration; the incorporation of a better specification of equilibrium-corrections; and by extending the sample by 8 years. The results suggest strong housing market effects consistent between inflow and outflow equations: houses/head, price levels, expected appreciation, and downside risk. We have modelled gross inand out-migration, as well as net migration of working-age individuals. We have also examined migration patterns for three age groups, - those aged 15 to 24, those aged 25 to 44 and the 45 to 64 year old group. We find that as age increases, so house prices matter somewhat more relative to earnings. Among other findings, the unemployment effects in London are far weaker for the over 44s than for the 15 to 24 and 25 to 44 age groups. Previous research either ignored house prices altogether, or only included level effects. Leaving out housing market effects typically results in misspecified models in which labour market effects are estimated as being weak or even perverse in direction. Net and gross flow equations are broadly consistent, - though, - for example, the downside risk effect is estimated to be more powerful in the outflow equations than the inflow equations, and the speed of adjustment of outflows is smaller than of inflows. Though every regional in-migrant is an out-migrant from another UK region, one should not expect the two equations to be exact mirror images of each other. 4 Contiguity matters - we find that, as economic incentives suggest, housing market comparisons with contiguous regions are more important than with the average of all regions. Given the commuting option, people may choose to live in regions with somewhat lower housing costs and commute to the contiguous region with higher 4 This is because both are weighted by the working age population of the region rather than consistently as an average of populations in the respective regions of origination and of destination. house prices for work. Indeed, we find some evidence that good labour market opportunities in contiguous regions have a considerable influence on migration to and from a region. A priori it seems likely that regions such as the West Midlands, entirely surrounded by other regions, would experience stronger contiguity effects than regions such as the South West. Our model builds in such variations in connectedness but the estimates of the basic economic responses appear to be quite robust to whether we assume connectedness is uniform or varies across regions. However, there is clear evidence of some heterogeneity in behaviour: migration to and from Greater London and the greater South East is more responsive to house prices, and has somewhat different responses to unemployment, from other regions. Greater London s migration responds between two and three times as strongly as other regions to the level of housing relative to population. As we saw, this explains much of the variation in net regional migration into Greater London, and suggests an important crowding out effect on this migration from international migration The evidence is that the unemployment rate appears to be more relevant for migration than the employment rate. In the 1990s, regional differentials in the unemployment rate narrowed much more than differentials in the employment rate, reflecting the higher levels of disability benefits claimants and other non-participation in the poorer regions. The finding suggests that such claimants and early retirees are less likely to be migrants or to respond to economic signals. While the micro-econometric literature has emphasised the role of tenure on mobility, we could identify no significant and plausibly signed effects, for example from regional social housing tenure rates on migration rates. This is probably because such effects are captured by the region fixed effects and the region time trends, and also because our migration rates are scaled by the sum of UK gross regional migration rates, which will capture aggregate variations in mobility due to tenure changes at the UK level. 5 Provided one remembers that the migration equations are only one link in an interconnected 5 It is also possible that shifts over time in tenurespecific mobility rates found by Hughes and McCormick (2000) may have made it hard to find stable coefficients. Migration within England & Wales 16 July 2005

system of housing, labour market and demographic responses, the model can be used to throw light on some questions. Table 3 summarises the estimated long-run responsiveness of net migration in the North, Greater London and the total South East (labelled South). For example, suppose house prices in the total South East (excluding London), were 1% higher relative to the average for Great Britain and the South East s contiguous regions (or relative earnings were 0.5% lower), and other factors, including expected relative appreciation and downside risk were unchanged, net migration of working-age people relative to the working-age population would drop by 0.20 per thousand compared with an average net annual flow for 2000-3 of 1.93 per thousand. A permanent rise of 1% in the expected rate of relative house price appreciation in the South East would offset 60% of this effect. A fall of 1% in London s housing stock relative to working age population, holding house prices, earnings and unemployment rates constant (eg because of changes in international migration), would decrease net regional migration rates into London by 0.78 compared to -10.73 per thousand on the average for 2000-3. More people in London would raise relative house prices, but if expected relative house price appreciation also rose, the two effects could easily cancel out, at least in the short run. Our estimates find hardly any response for migration to the South East from relative unemployment rates. But net migration to Greater London is strongly affected by relative unemployment. For example, a 1% point rise in unemployment in London relative to Great Britain and London s contiguous region (the South East) decreases net migration by 0.87 per thousand. The short-run effect is even bigger than the long-run one. It is wise to be cautious about drawing policy inferences from one set of relationships in a complicated web, but some things stand out. These include the importance of housing, both mediated through prices and direct availability, for regional migration in the UK, and the central and different way in which regional migration to London operates. Both have interesting implications for analysing the effects of increased house building in different regions on the affordability of housing in different regions. Another finding of our research is the important role played by relative expected house price appreciation, as well as down-side risk, in explaining migration. This helps to explain why the greater South East, for example, continued to attract net migrants when house prices were rising very strongly relative to other regions, even when they had become expensive relative to earnings. In turn, this helps a little in explaining why relative regional house prices in England and Wales have such sustained periods of appreciation. Table 3 The Estimated Responsiveness of Net Migration North Greater London South Net working-age migration per 1000 population average 2000-2003 -0.49-10.73 1.93 1% rise in relative house prices -0.06-0.10-0.20 Estimated longrun change in rates 1% point rise in relative unemployment -0.34-0.87 0 net migration 1% rise in relative earnings +0.12 +0.21 +0.44 in 2003 1% rise in relative working-age population to housing stock ratio -0.32-0.78-0.32 Notes: The estimated long-run effects are derived using the net migration parameter estimates in Table 2. Feedback effects from migration to house prices etc have been ignored and the rises in relative house prices etc are assumed to be permanent. July 2005 17 Migration within England & Wales

References Barker, K. (2004) Delivering stability: securing our future housing needs, H.M. Treasury. Borjas, G. (2003) The labour demand curve is downward sloping: re-examining the impact of immigration on the labour market, Quarterly Journal of Economics, 118, 4, 1335-1374. Bover, O, Muellbauer, J. and Murphy, A. (1989) Housing, Wages and UK Labour Markets, Oxford Bulletin of Economics and Statistics, 51(2), March, pp.97-136. Cameron, G. and Muellbauer. J. (1998) The Housing Market and Regional Commuting and Migration Choices, Scottish Journal of Political Economy, vol. 45(4), September, pp. 420-446. Dustmann, C., Fabbri, F., Preson, I. and Wadsworth, J. (2003) The Local Labour Market Effects of Immigration in the UK, Home Office online report 06/03. Gordon, I. (1975) Employment and Housing Streams in British Inter-Regional Migration, Scottish Journal of Political Economy 22(2), June, pp.161-77. Gordon, I. (1990) Housing and Labour Market Constraints on Migration across the North-South Divide, in J Ermisch (ed), Housing and the National Economy, Aldershot: Brookfield, pp.75-89. Gordon, I. and Molho, I. (1998) A Multi-stream Analysis of the Changing Pattern of Interregional Migration in Great Britain, 1960-1991, Regional Studies, 32(4), pp.309-323. Harris, J. and Todaro, M. (1970) Migration, Unemployment and Development: a Two-Sector Analysis, American Economic Review, 60, 126-142. Hatton, T. and Tani, M. (2003) Immigration and Inter-Regional Mobility in the UK, mimeo. Hornsey, D. (1993) The Effect of Computerization of the NHS Central Register on Internal Migration Statistics, Population Trends 74, pp. 34-36. Hughes, G. and McCormick, B. (1987) Housing Markets, Unemployment and Labour Market Flexibility in the UK, European Economic Review, 31(3), April, pp.615-41. Hughes, G. and McCormick, B. (1991) Housing Markets, Unemployment and Labour Market Flexibility in the UK, in G de Menil and R J Gordon (eds), International Volatility and Economic Growth: the First Ten Years of the International Seminar on Macroeconomics, Amsterdam, London and Tokyo, Elsevier Science: New York, pp.83-109. Hughes, G. and McCormick, B. (2000) Housing Policy and Labour Market Performance, ODPM. Jackman, R. and Savouri, S. (1992a) Regional Migration versus Regional Commuting: The Identification of Housing and Employment Flows, Scottish Journal of Political Economy, 39(3), August, pp.272-87. Jackman, R. and Savouri, S. (1992b) Regional Migration in Britain: An Analysis of Gross Flows Using NHS Central Register Data, Economic Journal, 102(415), November, pp.1433-50. Kiel, K. (1994) The Impact of House Price Appreciation on Household Mobility, Journal of Housing Economics, 3(2), June, pp.92-108. McCormick, B. (1997) Regional Unemployment and Labour Mobility in the UK European Economic Review, 41(3-5), April, pp.581-89. Molho, I. (1982) Contiguity and Inter-Regional Migration Flows in Great Britain, Scottish Journal of Political Economy 29(3), November, pp.283-97. Muellbauer, J. and Murphy, A. (1991) Regional Economic Disparities: the Role of Housing, in Bowen, A. and Mayhew, K. (eds) Reducing Regional Inequalities, NEDO and Kogan Page, London. Muellbauer, J. and Murphy, A. (1997) Booms and Busts in the UK Housing Market, Economic Journal, 107(445), pp. 1701-1727. Migration within England & Wales 18 July 2005

Appendix Table A1 Some Summary Statistics and Regression Diagnostics for the Net Migration Equations Region Mean of Dependent Variable Std Dev of Dependent Variable Equation Std Error R Squared DW Statistic North -0.206 0.120 0.045 0.85 2.24 Yorkshire & Humberside -0.054 0.107 0.061 0.67 1.54 West Midlands -0.226 0.066 0.043 0.57 2.08 East Midlands 0.218 0.108 0.065 0.63 1.61 South 0.289 0.156 0.034 0.95 2.12 Greater London -0.465 0.269 0.094 0.88 2.40 South West 0.563 0.149 0.078 0.71 1.64 Wales 0.154 0.140 0.068 0.78 1.71 Note: The dependent variable in each equation is the change in the weighted net migration rate of the workingage population in that region. The North and South regions are the sums of the North East and North West, and South East and Eastern Government Official Regions respectively. July 2005 19 Migration within England & Wales