An expression commonly used by economists about the ability of value-added tax to fetch revenue is that VAT (value added tax) is a money-spinner or a money-machine.

I have examined this claim empirically and found that it is a myth. The data available is not comprehensive but adequate to draw or demolish a conclusion.

There are eight countries in our study (VAT revenue as a proportion of GDP) where the growth of VAT revenue as percentage of GDP has been negative.

One of them is Brazil, about which the data is not exactly comparable to the above countries but the conclusion is the same, that there has been a negative growth over a period of time.

In Brazil, in 1968, VAT revenue, as a percentage of total tax revenue was 51.03, while in 1985 it became 21.91, and in the period in between, it gradually fell.  In comparison, income tax rose from 21.8 per cent in 1968 to 58.97 per cent in 1985.

On the other hand, 11 countries in our study have performed satisfactorily (3 per cent growth and above).

However, we have to consider that in the European countries, the revenue from excise on cigarette, beer, wine, spirit and gasoline was a considerable percentage of GDP.

In 2001, the percentages were the following: Denmark 5.7, Portugal 4.8, Luxembourg 4.8, Greece 4.7, Finland 4.7, Ireland 4.6, the UK 4.1, Sweden 3.7, Italy 3.5, the Netherlands 3.3, France 3.0, Spain 2.8, Austria 2.7, Germany 2.6, and Belgium 2.4.

The average for the European Union was 3.8. And the percentage of excise revenue to the total tax revenue was 9.6 per cent as an EU average.

Therefore, the performance of VAT in EU countries is to be taken as 9.6 per cent less since that is the share of excise in what goes as VAT revenue.

Although some of these 11 countries have performed well after VAT was introduced and can be categorised as good, the corresponding percentage of growth in the years previous to the introduction of VAT are also very high.

We have the figures of some countries, which are comparable. In the case of Korea, the previous rate of growth was 23.63 per cent and the present rate is 8.74 per cent.

In the case of Senegal, the previous rate was 9.01 per cent and the current rate is 4.53 per cent. For Nicaragua the past rate was 14.5 per cent and the current  is 9.82 per cent.  For Mexico the past rate was 6 per cent and the current is 8 per cent.

Thus, in many countries where comparable data are available we find that the previous rate was even more than or almost equal to the present rate.

If the previous rates of other countries were available, the situation most probably would have been similar.

In 12 countries, where previous rates are available, they are very high.  In some cases, like in Israel, Costa Rica and Peru, the previous rates were higher and now the rates are negative.

In the case of Germany, where VAT was introduced in 1968, there is more comprehensive data available and the result shows that VAT revenue as a percentage of total tax revenue was 17.1 in 1970 and it decreased to 14.6 in 1975, 15.1 in 1978 and ultimately reached 17.9 in 2001, a figure which is not much better than it was 31 years ago.

VAT revenue as a percentage of GDP also increased from 5.6 in 1970 to only 6.6 in 2001. This also after the VAT rate itself was increased from 4 to 11 to 13, and then to 14, 15 and 16 per cent.  When the VAT rate itself grew four times, it is only natural that the revenue from VAT would increase.

The increase in revenue from VAT cannot be attributed to the efficiency of VAT only.  Moreover, what is true of Germany would also definitely be true about the UK and other countries.

Although the chronological figures are not readily available, the fact is that in several countries, the rates have increased over time, for example, in Japan from 3 to 5 per cent. The increase in VAT rate in Germany cannot be a singular phenomenon.

There are some other studies on regional basis, which are relevant  (Revenue difference between VAT and predecessor sales tax by region).

In this table, there are no comparable figures available for revenue trends before VAT was introduced.  In South America and also in Europe we have seen from the first table that the rate of revenue growth has not been uniform.

Taking the figures as they are, we find that there have been negative growth in central Europe and practically no growth in north Africa and west Asia.

The growth has varied between 1.88 and 1.96 per cent. If the sales tax to GDP ratio was 10 in central Europe, to use a purely hypothetical number, after VAT this fell to 8.12.

In Small Islands, the growth is impressive but they are not quite significant to come to a general conclusion. This table provides only mixed evidence about revenue, which even the author, Liam Ebrill, admits.

The conclusion is that those who call VAT a money-spinner have not taken into account various factors, namely (i) the percentage  growth in previous years; (ii) increase in the rate of tax;  (iii) the fact that in many countries the growth is negative; (iv) the EU average itself is marginally positive.

EU's yearly average is only 0.25 percentage, which means that Europe as a whole has not fared so well; and (v) other taxes, namely, income and security taxes have grown much more than VAT.  A combination of sales tax and excise is a better money-spinner than a combination of VAT and excise.