The retail sector has been responsible for a large share of the relatively poor performance of the European Union compared to the United States, and policy-makers and academics have focused on regulation as one potential reason. Policy proposals such as The Services Directive aim to deregulate these markets and thus increase competition across Europe.
Recent work suggests that the adoption and use of technology was an important contributor to the US productivity acceleration of the late 1990s. The literature emphasises the importance of entry for productivity growth in US retail, enabling the more rapid introduction of information and communication technologies (ICT) in newer stores and leading to reallocation from less productive to more productive stores.1
In contrast, UK productivity growth has largely come from incumbents. As in other EU countries, the UK retail sector has experienced relatively low productivity growth and concentrated market structure. UK competition authorities have repeatedly investigated the supermarket industry. Attention has focused on land use regulation as one of the root causes of this poor performance. Previous academic work looking at the French retail sector (Bertrand and Karamtaz 2002) found that regions with more restrictive planning regulation had lower levels of employment. Do UK planning regulations inhibit entry that drives productivity growth?
UK planning regulations
The UK reformed its planning regulations in the mid-1990s. The new regulations severely restricted the ability of firms to build new large-scale out-of-town supermarkets. An influential report by McKinsey (1998) highlighted these reforms to planning regulations as one of the major policy issues affecting productivity in the UK, and in response the UK has undertaken a number of reviews of the land-use planning system in England, with particular focus on the impact on productivity.
Despite this interest, there is relatively little sound empirical evidence on the impact of land use regulation. Work commissioned by the Office of the Deputy Prime Minister (ODPM) has suggested that land use regulation has had a large impact on entry of new large stores, mainly by comparing entry rates before and after the reform of 1996, as shown in Figure1. Before this major reform, approximately similar numbers of small and large stores were opened, while after the reforms, there was a rapid rise in the number of smaller format stores opened.
However, the problem with such a before-and-after comparison is that it does not account for other factors that may also have affected the shift towards smaller store formats - for example, other commercial and strategic reasons that UK supermarket firms may have moved into the smaller store format.
Planning regulation and market structure
In a recent paper, Griffith and Harmgart (2008), we consider the impact that planning regulation has had on market structure in the UK supermarket industry, taking into account some of these other factors. We show that when we do not take these other factors into account, we incorrectly attribute other contemporaneous changes to the regulatory reforms. We use variation in the way that the regulations were implemented across local authorities to identify the effect of regulation on market outcomes. We compare the impact of regulation across store types that were and were not affected by the regulation.
We model firms as profit maximising. We assume consumers demand the bulk of groceries in "one-stop" and the residual in a "top-up" shop, a common way to model consumer demand in the economics literature. We assume that firms make decisions about entry into the large store format independently of the number of small stores, whereas the decision to enter into the small store format takes the number of large stores as a given. We think this is a realistic assumption, motivated by the one-stop versus top-up model of shopping behaviour. We use information on the geographic concentration of economic activity in town centres in the UK to define local markets.
Our empirical estimates suggest that profits are higher in regions where there is a larger population, and that the factors that affect variable profits suggest that these are higher for larger stores (above 15,000 square foot) in areas with lower unemployment, further away from another town centre, with more density of office floor space. Unsurprisingly, higher retail land values are associated with lower profits. Large stores are more profitable in the South of England, while smaller format stores are more profitable in more working class areas. For all stores, fixed costs are lower in the North.
Our findings show that when we do not control for these observable differences that affect fixed costs, our estimates of the effect of planning regulations is over twice as large as when we condition on these factors. Overall we find that planning regulation did have a statistically significant impact on the number of firms operating in a region (the market equilibrium), so it does represent a barrier to entry. A one percent increase in the approval rate of planning applications has a similar magnitude of effect (but opposite sign) as a one percent increase in the unemployment rate. To further quantify the economic impact of the reform, we use additional data to look at how the price of groceries vary with the planning regime. We find that restrictive planning regimes are associated with a small but statistically significant increase in food prices.
What role has regulation had in shaping market structure? Our results suggest that the impact of planning regulation may have been overstated. While it has had an impact, the overall impact seems to have been relatively small in terms of the price that consumers face.
Bertrand, M and F Kramatz (2002). “Does Entry Regulation Hinder Job Creation? Evidence from French Retail Industry,” The Quarterly Journal of Economics, November, 1369-1413
Griffith, R and H Harmgart (2008). “Supermarkets and Planning Regulation” CEPR Discussion Paper 6713.
McKinsey (1998). "Driving productivity and growth in the UK economy" McKinsey Global Institute, October.
1 While it is likely that ICT usage is higher and more effective in newer shops, it is not clear whether a market with fewer large out-of-town stores or more small in-town-stores will lead to more entry and exit. It may also be that ICT is easier to adopt in larger out-of-town stores. For discussion of these issues see a number of working papers from the Groningen Growth and Development Centre and references therein.