This evidence is submitted on the part of Professor Peter Gripaios, Dr Eric McVittie and Dr Steven Brand of the South West Economy Centre, University of Plymouth, Drake Circus, Plymouth, PL4 8AA.

  1.  It is not clear that there are consistent differences in regional growth rates for per capita income. The stable long-run pattern of regional disparities identified by, for example, HM Treasury (2001) suggests only relatively minor and transitory differences in growth rates of income per person[1]. This may suggest a "steady state" pattern of regional disparities. A successful regional policy must be based on an improved understanding of the mechanisms supporting and maintaining this steady state.

  2.  The key issue is, then, differences in levels of GDP per head between the UK regions, rather than differences in growth rates. These levels differences reflect underlying weaknesses in some regions with regard to, for example, human capital, R&D, level of technology, access to markets and agglomeration economies.

  3.  In this regard the most significant differences are between London (particularly Inner London) and its hinterland, and the rest. On a workplace basis, London's GDP per capita during 1999 was almost 1.5 times that of the UK as a whole. Inner London has the highest level of GDP per capita of all EU NUTS2 regions (at around 2.4 times the EU15 average during 1998). [2]London is a world city and we should not be surprised by these findings.

  4.  Some other large UK cities outside the South East are doing very well relative to others and to more rural areas. These are mainly the provincial capitals eg Bristol, Leeds & Edinburgh. Some such areas have seen marked increases in GDP per head relative to the UK average. They have been winners at the expense of other towns and cities within their region and sometimes over a wider area. For example Leeds has become the main business services centre (outside London) for the whole of the north of England. Bristol has taken on a similar role for the South West and to some extent for South Wales. The rationalisation and concentration of financial capital and regional office functions has been an important factor. This has, however, been supported by the effect of the setting up of regional offices of Government Departments over the long term and more recently of Government Offices for the Regions. Since such locations are also the main centres of Higher Education outside of Oxbridge and London, it is not hard to see that they have a lot of cards stacked in their favour in an increasingly, "knowledge-based" economy.

  5.  Assuming that we are really interested in reducing regional disparities, it would be helpful to have more confidence in the quality of regional data. Too much of that is derived from national surveys and recording systems not specifically designed to produce high quality regional data. The processes by which such data are attributed to the regions are opaque and give rise to unknown (but probably substantial) errors. Plans within ONS to improve the quality of regional data are to be welcomed, but probably do not go far enough. Investment in the collection of quality regional data is essential if we are to have a sound "evidence base" for decisions on regional policy priorities.

  6.  The quality of GDP/GVA and related productivity measures (GDP per capita, GDP per job, GDP per hour) for the UK regions is highly questionable in our view, and that for sub-regional measures even more so. ONS have acknowledged problems with the recently revised GVA estimates incorporating Annual Business Inquiry Financial Analysis (ABI/2) data, which have subsequently been withdrawn. Data problems are, however, much more widespread than this. The quality of sub-regional information is especially important given the mixed nature of many of the Government Office Regions that do not correspond to functional economic regions. Data on the South West as a whole, for example, is of little worth in understanding regional disparities, disguising at it does substantial variations between sub-regions. We have little confidence that sub-regional data supports detailed descriptive analysis of disparities, let alone rigorous analysis of the causes of sub-regional differences.

  7.  Even if the figures were accurate, care needs to be taken as to how they were interpreted. At sub-regional level Cornwall is often described as the poorest county in England because GDP per head is only 65 per cent of the UK average. But economic factors such as a high percentage of low wage employment are only part of the explanation. Cornwall also has low measured GDP per head because it has large numbers of retired people and large numbers of commuters to Plymouth in Devon where their output is counted. If we look at other measures such as household disposable incomes, Cornwall looks much better off, certainly relative to other UK areas which qualified for Objective One and indeed relative to some which did not.

  8.  The importance of cities as engines of economic activity within their surrounding regions is well recognised. The very existence of large cities suggests substantial economies from the spatial concentration of economic activity, including savings in transport and communications costs for products (Krugman, 1991), intermediate inputs (Venables, 1996; Ciccone & Hall, 1993), and ideas (Lucas, 1988, Rauch, 1993)[3]. Large urban labour markets reduce search costs for firms and workers and so tend to improve the quality of labour market "matches" (Wheeler, 2001) and may contribute to skills acquisition (Glaeser & Mare, 2001)[4]. Large cities will tend to attract high skill workers, for whom the benefits from large labour markets are more pronounced. Large cities also tend to be more diversified and more innovative (Duranton & Puga, 1999). [5]Geographical factors such as the absence of (access to) large population centres will place constraints on a region's productive capacity, making it difficult for regions without large and dynamic cities to match the performance of more urbanised areas.

  9.  Doing something about regional disparities should not involve levelling down. This means that the central role of London and the more dynamic regional cities should be acknowledged, supported and strengthened. This can be accompanied by efforts to strengthen the links between those cities and their surrounding areas.

  10.  Stable differentials among the UK regions have apparently persisted in the face of massive industrial, organisational and occupational restructuring, revolutions in communications and transport technology, and large shifts in all areas of government policy. This suggests that powerful forces are in play tending to perpetuate these differentials. This is not to imply that nothing can be done to alter this pattern, but should caution against unrealistic expectations. It is also important that, so far as possible, policy works with rather than against these forces. Otherwise expensive and temporary short-term fixes may be the best that can be achieved.

  11.  It is very difficult to decide whether current policies are having an impact given data quality and long time lags involved in both policy impacts and the collection of relevant indicators. There are also important issues of "noise" and of whether regional policies can be considered a success if they have at least stopped things from getting worse. As for past policies, the same points apply. Moreover, it is perhaps unfair to judge past policies against present criteria.

  12.  It is too early to attempt to evaluate the success of either Regional Assemblies or RDAs. In our view, it will be difficult for Assemblies to take a regional view, especially in regions such as the South West, where there is little feeling of regional identity. Tribalism is strong and there is little chance that, for example, the citizens of Bath let alone those of Plymouth, Penzance or Exeter will willingly accept the importance or hegemony of Bristol.

  13.  RDAs might make a difference on the margin but only if adequately funded, if they can attract quality staff and if they are able to focus on key and deliverable objectives. They will need to do more than they have so far to enlist the co-operation and support of the private sector.

  14.  The movement of Central Government staff from London could have an impact. There is no reason for so many civil servants to be in the capital. As in the private sector, many more routine functions could be transferred to lower cost locations. This would also reduce public expenditure and congestion in London, ideally helping the competitiveness of the latter. Other, higher-level functions may also be usefully relocated to provincial cities. In our view, it is reasonable to argue that the concentration of public and semi-public expenditure in specific localities may well have had a catalytic effect on the private sector. In Bristol, the concentration of MOD Procurement at Abbey Wood has pumped millions into the local economy and helped to attract private sector activities to the North Bristol/South Gloucestershire area.

  15.  An important issue is whether any transfer of functions should be spread thinly or concentrated in a small number of locations.

  16.  The same point applies to Government expenditure more generally, and the private expenditure which the Government can influence on air and other transport, higher education, research and development and telecommunications. There may be benefits from focusing major infrastructure projects so as to capitalise on the growth of the most successful provincial centres, to try to get a Barcelona-type effect. Certainly, we do not see that everyone can be a winner and from the perspective of an unemployed resident of Camborne or the Rhondda Valley, a job in Bristol or Cardiff may be a more attractive proposition than one in Bordeaux or Munich.

  17.  In any event, it would seem important to improve transport links between the most successful urban areas and their hinterlands as well as between London and the regions.

2   European Commission (2002) Second Report on Economic and Social Cohesion-Statistical Annex. Back

3   Paul Krugman (1991) Geography and Trade, MIT Press; Anthony Venables (1996) "Equilibrium locations of vertically-linked industries", International Economic Review ; Antonio Ciccone & Robert Hall (1993) "Productivity and the density of economic activity", American Economic Review, Robert Lucas (1988) "On the mechanics of economic development", Journal of Monetary Economics, James Rauch (1993) "Productivity gains from geographic concentration in cities", Journal of Urban Economics. Back

4   Christopher Wheeler (2002) "Searching, sorting, and urban agglomeration", Journal of Labor Economics; Edward Glaeser & David Mare (2002) "Cities and skills", Journal of Labor Economics. Back

5   Gilles Duranton and Diego Puga (1999) "Diversity and specialisation in cities: Why, where and when does it matter?", Centre for Economic Performance Discussion Paper No. 433. Back