MINI CASE 15.1

Once you have data that allows the plotting of scatter diagrams, you will also be able to calculate a correlation coefficient. However, you still need to give an interpretation to this figure and argue the case that it does make sense. If you look at sufficient data, you will eventually find a correlation of some sort. The question then becomes ‘does this correlation give a reasonable description of a possible relationship?' If you use ‘correlation' as a key word in a newspaper search you will find lots of examples of how a case is being made on the basis of correlation.

The fact that a correlation can be shown to exist is often used to win an argument. You still need to decide whether the correlation presented is indicative of a cause and effect.

Corporation tax cuts do not stimulate growth

Prompted by the coalition government's promise to stimulate growth and cut corporation tax from 28 per cent to 23 per cent in 2013, Richard Murphy, on behalf of the TUC, has produced a report that claims that the links between tax rates and employment and growth are weak. Based on OECD sample data that were similar to the UK he says that although the data cannot prove causal relationships it does indicate that only 7 per cent of all variation in growth and employment could be explained by tax rates. In this case Murphy suggests, ‘to change tax rates is a very poor way to stimulate the economy'. The TUC report acknowledges that the analysis does run counter to previous work by the Organization for Economic Co-operation and Development (OECD) which concluded that lower corporation tax rates does stimulate economic growth.

In criticizing the report an Oxford University tax expert, Mike Devereux, pointed out that it ignored some of variables that might have had relevance in this instance and added ‘To say corporation tax does not affect growth is just ignoring quite a lot of academic literature'.