• This article presents data on the stock of VAT-based businesses, registrations and deregistrations for Scotland and the UK for the period 1980-96. It discusses the coverage of the data and certain discontinuities in the series.
• The total number of VAT-registered businesses in Scotland increased in each year from 1980 until 1991, with particularly strong growth in the stock in the period 1988-90. The early 1990s were a more difficult period in terms of the VAT stock, with net falls each year between 1992 and 1995.
• A similar pattern emerges at the UK level. However, UK growth was stronger than Scottish growth during the 1980s, and the decline in the stock began a year earlier in the UK, in 1991.
• At a sectoral level, the retail sector has seen a long-term decline in the stock of VAT-registered businesses in both Scotland and the UK. The sectors which grew most strongly in Scotland over the period 1994-96 were finance, business services and public administration. Over the earlier period, finance and other services also grew strongly.

Introduction

VAT-based data² are often used as indicators of trends in the business stock. This is because, apart from certain sectoral exclusions, the information is comprehensive for businesses with a turnover in excess of the minimum registration threshold (£49,000 from 1 December 1997). It covers different types of legal form, such as sole trader or partnership, as well as individual firms. It is also available over a reasonably long time period and at a spatially disaggregated level. Its main drawback for this purpose is that it excludes the smallest businesses which are not required to register, unless they choose to. Estimates from the SME Statistics Unit at DTI suggest that, at start-1996, 43 per cent of all UK enterprises were registered for VAT.

Although VAT registrations and deregistrations are used to proxy business births and deaths respectively, it is important to note that registrations and deregistrations are not synonymous with business births and deaths. In particular, registrations will include businesses which have been trading for some time and have just crossed the threshold, and will exclude new start-ups which are below the threshold. Deregistrations will occur because turnover falls below the threshold, and for other reasons such as a change of ownership, or change of legal status, as well as when a trader unambiguously goes out of business. During 1995, only 55 per cent of businesses in Scotland which deregistered did so because they ceased trading. 18 per cent deregistered due to a change in legal status, 17 per cent did so because their turnover dropped below the threshold, and the remaining 10 per cent did so for other reasons³.

VAT-based Data for Scotland

VAT-based data for Scotland are available annually from 1980, but there are discontinuities at 1991 and 1993. Large increases in the threshold in 1991 and 1993 mean that estimates for 1980-91, 1992-93 and 1994-96 are not entirely comparable. Furthermore, the estimates for 1980-93 count VAT reporting units while figures for 1994-96 count whole VAT registered enterprises. An enterprise is a legal unit, person, or group of people producing goods or services under their own control and with their own legal identity. A branch or office of a larger organisation is not in itself an enterprise. There may be more than one VAT unit within an enterprise. Thus the enterprise corresponds more closely to the concept of the "firm" or "business". A further discontinuity when examining industrial sectors is that for 1980-93 the estimates are broken down by VAT Trade Classification (VTC) while for 1994-96 they are broken down according to the 1992 Standard Industrial Classification (SIC92). Finally, the earlier data are available for the old local authority areas while the 1994-96 data are available for Scottish unitary authorities.

Table 1 sets out the data for Scotland for the period 1980-96, with the discontinuities indicated.

TABLE 1 HERE

Although the figures are not entirely comparable over the whole period, it is reasonable to look at the net change in any particular year and the relationship between registrations, deregistrations and initial stock. Table 1 shows that the total number of VAT registered businesses in Scotland increased in each year from 1980 until 1991. The rate of change increased in the period 1988-90, as a result of an increase in the registration rate coupled with a constant deregistration rate. The table also indicates that the first half of the 1990s was a more difficult period for the business sector than the 1980s. In 1991, the rate of increase in the stock of VAT-registered businesses fell sharply, as a result of a fall in new registrations, combined with a lesser increase in deregistrations. This was followed by reductions in the total stock⁴ in each year between 1992 and 1995, as deregistrations rose to a peak of 13,600 in 1993 while registrations fell.

The stock of VAT-registered businesses appears to have been more closely linked to the business cycle during the late 1980s and early 1990s than in the early 1980s, when the VAT stock continued to increase during a much deeper recession. However, the VAT stock has turned around more slowly than GDP during the recent upturn as shown in Chart 1.

CHART 1 HERE

In this context it is useful to consider the industrial breakdown of the VAT stock and of registrations and deregistrations. The analysis is complicated by the change from VTC to SIC 92. Nevertheless, some interesting points emerge. Table 2 looks at the more recent period, 1994-96, where SIC92 is the basis of the industry breakdown, while Table 3 looks at the earlier period using the VTC.

TABLE 2 HERE

TABLE 3 HERE

The retail sector has seen an inexorable decline in the stock of VAT-registered businesses. This is the only sector to have seen a decline in every year. This reflects the move from small retail outlets to larger ones, and the trend towards large out-of-town shopping centres. During the early 1980s, the only other sector which experienced a declining VAT stock was transport. In contrast, most sectors experienced falls in the VAT stock in several years between 1991 and 1996. During the 1991-1993 period, finance and other services continued to grow significantly and the wholesale sector also held up well until 1993. During the 1994-1996 period, the financial intermediation, business services and public administration sectors remained buoyant. The largest falls occurred in wholesale, retail and repairs. Manufacturing and construction also fell.

It is difficult, on the basis of these tables, to argue that the more service-related nature of the 1990s recession was responsible for the overall trend in the VAT stock. Many of the sectors which held up well were in the service sector, while non-service sectors suffered. Among the service sectors, wholesale, retail and repairs was the most badly affected. Table 4 looks at the service sector as a whole compared with the rest of the economy and illustrates the comparable strength of the service sector over both the 1980s and 1990s.

TABLE 4 HERE

VAT-based Data for the UK

A similar picture emerges at UK level, although the decline in the stock begins a year earlier in 1991. Table 5 provides the UK data and Tables 6 and 7 provide an industrial breakdown for the UK.

TABLE 5 HERE

TABLE 6 HERE

TABLE 7 HERE

As in Scotland, the retail sector has witnessed significant decline in terms of the VAT stock. The pattern across the other sectors is also broadly similar, although the catering sector appeared to suffer an earlier downturn at UK level than in Scotland, and the extent of the decline in the VAT stock in the construction sector has been more pronounced at UK level. The finance sector in Scotland has also seen stronger growth in most years.

Scotland/UK Comparisons

Net Changes in the Stock of Companies

To compare Scotland with the UK as a whole, it is useful to look at the net change as a percentage of the initial stock as shown in Table 8.

TABLE 8 HERE

This illustrates a well-known feature of the business stock in Scotland: that it failed to grow as rapidly during the 1980s as the UK stock of businesses did⁵. Between 1980 and 1990, the stock of VAT-registered units rose by 27 per cent in the UK but by only 19 per cent in Scotland. Since 1990, however, Scotland's relative position appears to have improved. The downturn in the stock of businesses did not start until a year later than in the UK and was less pronounced in the years 1992, 1993 and 1994. However in 1995 and 1996 the UK has again performed relatively better than Scotland.

Shift-share Analysis

A simple shift-share analysis highlights the differences in sectoral growth in Scotland and the UK. Because of the data discontinuities and the change in industrial classification, the analysis was carried out for the period 1980-92, providing an overview of the 1980s, and for the most recent period, 1994-96. The percentage change in each UK sector was applied to the Scottish stock in the base year. The results are shown in Tables 9 and 10.

TABLE 9 HERE

TABLE 10 HERE

This analysis confirms that, over the more recent period, the Scottish VAT stock has been more robust than the UK stock. The total stock would have been 509 or 4 per cent lower, had each sector experienced the same rate of growth as the UK sector. It also highlights those sectors where growth was slower than in the UK as a whole - manufacturing, wholesale and retail, transport and business services. Sectors which performed more strongly than their UK counterparts were public administration, finance, and education and health.

Over the earlier period, in aggregate, the reverse is true - the total Scottish VAT stock would have been more than 7,000 higher had each sector grown at the same rate as its UK counterpart. Finance, catering, agriculture and production performed relatively better in Scotland.

Unitary Authorities Data

The VAT-based data are also available by unitary authority within Scotland. Annex 1 gives the stock of enterprises, registrations and deregistrations by SIC92 for 1994-96 for each unitary authority. Prior to this, data are available for the old local authority regions by VTC.

Survival Rates of Businesses

Table 11 shows'survival rates' of VAT-registered businesses in Scotland. These show the proportion of businesses remaining registered for VAT at 6-monthly periods after their initial registration. Looking at survival after 12 months, survival rates were higher in the late 1980s, though there are signs that recent survival rates are improving.

TABLE 11 HERE

Table 12 shows the same information for the UK. Survival rates of Scottish VAT-registered businesses are in most cases better than for the UK as a whole. This may be related to the sectoral mix. As with births and deaths of businesses, caution must be used when considering these survival rates as proxies for the economy as a whole. These estimates are likely to over-estimate true survival rates because they exclude the very smallest businesses (those not registered for VAT) with the lowest survival rates.

TABLE 12 HERE

Conclusion

The VAT-based data provide a useful guide to movements in the stock of businesses and to the pattern of start-ups and closures across the UK and between sectors. Although registrations and deregistrations are not synonymous with business births and deaths, when taken together with other sources of information on business start-ups, failures and stock of businesses, the VAT data act as a valuable indicator of trends in business activity.

Annex 1

TABLE a1 HERE

TABLE a2 HERE

TABLE a3 HERE

TABLE a4 HERE