Indicators suggest that industrial output growth has weakened and that this trend will most probably continue in March and April.

Industrial output growth for February was announced. But first I want to talk about one “confusing” feature of industrial output growth figures. Generally, we assess industrial output with two methods. First, we make a year-on-year comparison, that is, assess the annual growth rates. Also, the Turkish Statistical Institute (TURKSTAT) has recently started to announce seasonally and working-day adjusted industrial output figures which allow month-on-month comparison.

Nevertheless, adjusted figures can sometimes be confusing. In January, seasonally and working-day adjusted industrial output decreased month-on-month by 3 percent. In February, industrial output picked up by 0.7 percent compared to January. This zigzagged trend is not incidental.

There are twenty six months between January 2010 and February 2012. In twelve of these, month-on-month industrial growth was negative, that is, industrial output showed a decline compared to the month before. Out of these, four were recorded in 2010 and seven in 2011, years in which the economy grew impressively by 9.2 percent and 8.5 percent, respectively. For example, industrial output shrank by 2.2 percent both in August and November 2011.

Any statistical adjustment method is based on filtering the original data and the original series might transform considerably depending on the filter. I think it would be useful to test occasionally whether or not these filters serve their purpose. Anyway, this is an issue that must be dealt by the TURKSTAT. So let me skip this and concentrate on annual output growth. Between March and November 2011, year-on-year industrial output growth had an average of 8.2 percent. Starting with December 2011, pace of industrial output growth started to decline as was anticipated. The ratios for December 2011, January 2012 and February 2012 were 3.7 percent, 1.5 percent and 4.4 percent, respectively.

In spite of fluctuations, these figures imply an evident downwards trend compared to March-November 2011 period. Month-on-month growth figures can also fluctuate, not always because of the filtering method (in fact, in our case the non-adjusted series is also fluctuating). In order to avoid this, we can check quarterly figures. Starting with the first quarter of 2011, annual pace of output growth has been declining: the rates were 14.4, 8, 7, 6, and 6.5 percent respectively. As of February 2012, annual industrial output growth was 3.2 percent when the figures for the preceding three months were taken into account.

We can revisit the outlook in 2007 to get a better intuition about the current performance. In 2006, Turkish economy grew by 6.9 percent. Average growth declined to 4.7 percent by 2007. Average industrial output growth in the same year was 7.3 percent. What is more, excluding December, the lowest rate of industrial output growth in 2007 was 4.3 percent. Please note that in a year when industrial output grew by 7.3 percent, the economy grew by 4.7 percent, a rate that was considerably below the previous year’s growth.

I am sorry for overwhelming you with statistics. Here goes the moral of the story: It is not possible to identify the trend of industrial growth solely on the basis of monthly changes or the trend of economic growth solely on the basis of industrial output growth. The discussion above and the changes in other indicators (the slowdown in banks’ credit supply, the leading indicators, the drop in the exports to Europe and the current state of the automotive sector) suggest that industrial output growth has weakened and that this trend will most probably continue in March and April as well.

This commentary was published in Radikal daily on 10.04.2012