Commentators often allude to the fact that the average house price is now a much larger multiple of the average industrial wage than in the mid-1990s, giving rise to worries about the sustainability of the rise in house prices. In 1994, the average industrial wage was less than £14,000 per annum and had risen to less than £17,000 by 1999.
However, average industrial earnings, or indeed any average of wages across the economy as a whole, may not the best benchmark for the sustainability of the housing market. Incomes are not evenly distributed across the working population so an average measure is a poor guide to the situation at the extremes of the distribution of income.
Since the number of transactions in the housing market in any one year is small relative to the working population, by definition only a small segment of the population is active in the market in any given year, and their incomes may be nowhere near the average wage in the economy.
The number of transactions involving borrowings in the housing market was less than 80,000 in 1999 (up from below 50,000 in 1995), which is small in relation to a workforce of 1.7 million and a housing stock of well over one million homes.
The Revenue has recently published its annual statistical report relating to taxes and incomes in the year 1997/98. This enables a comprehensive picture to be shown of the distribution of income among taxpayers.
If we organise the number of taxpayers into 10 ranks or deciles of equal size by ascending level of income, we can see how, say, the bottom 10 per cent of taxpayers fare over time compared to the top 10 per cent. D1 is the lowest decile or 10 per cent of taxpayers and D10 the top 10 per cent.
The chart shows the average new house price expressed as a multiple of the average after-tax income of each decile from five to 10 - i.e. the top 60 per cent of earners. The bars in the chart show the multiples; the percentage rise in income for each decile is shown by the line in the chart which refers to the right hand scale.
We can see that, between 1993/4 and 1997/8, new house prices as a multiple of after-tax income in decile five rose from six to nine. Average after-tax income in this decile rose 15 per cent over this period from £9,500 to just under £11,000.
However, for decile 10, the multiple only rose from 1.7 times to 2.4 times because the average after-tax income in this decile rose by almost 24 per cent from £33,000 to £41,000. The aggregate after-tax income of decile 10 in 1997/98 was £4.8 billion, or 27 per cent of the aggregate after-tax income of all taxpayers.
The explanation for the large disparity in income growth between the top decile and the others is mainly that the income tax data is recorded on a case basis. Thus married couples both earning are one tax case and are one decision-making unit in the housing market. When the second spouse begins to work, obviously the average income of that tax case grows dramatically.
We know from labour force data that married women have added significantly to labour force growth since the early 1990s. In addition, the trend towards increasing cohabitation and single people joining on a pure business footing to buy houses means that the combined incomes directed at the housing market are not confined to married couples.
By 1999, not surprisingly, 75 per cent of loans to two-income borrowers were to those earning more than £30,000 per annum gross.
Thus the housing market is underpinned by the distribution of income towards the top decile brought about mainly by increased female participation in the workforce. The simple average measure of wages misses this picture. The relevant economic units active in the housing market are based on combinations of income in relation to which the house price has not risen dramatically during the second half of the 1990s.
Eunan King is senior economist at NCB
