“People who love sausage and people who believe in justice should never watch either of them being made.” So said Otto von Bismarck, the great German statesman of the 19th century, according to legend.

I may add “statistics” to this. That’s what I have been thinking about ever since the Turkish Statistical Institute (TÜİK) announced the 2017 first quarter growth figures. According to TÜİK, in the first quarter of 2017, Turkey grew 5 percent. Does that mean that there is a surge in growth in the Turkish economy in 2017 compared to 2016? I doubt it. I tend to think that the surge in the growth rate in the first quarter of 2017 is somehow related to the surge in the Turkish Lira’s inflation rate. It is all in the mechanics of how the chain linking of national accounts – TÜİK’s new estimation method - works.

Let me give you three numbers about the Turkish economy. Then I will come to the mechanics of sausage-making. First, Turkey still has a highly dollarized economy. It’s the original sin, and was first defined in 1999 by Barry Eichengreen and Ricardo Hausmann, “in which the domestic currency cannot be used to borrow abroad or to borrow long-term even domestically.” That is why out of necessity our economies are highly dolarized and open to exchange rate shocks.

Just look at the share of FX bank deposits in Turkey’s system. The ratio of FX bank deposits to total bank deposits was around 57 percent before the IMF rescue program in 2001. By 2010 this had declined to around 29 percent. But in March 2017 it had again risen to 45 percent. So Turkey is still a highly dollarized economy.

Secondly, the rate of depreciation of the lira was around 25 percent in 2015 and 20 percent in 2016. Overall, the rate of depreciation was around 50 percent. In a highly dollarized economy, in the case of an exchange rate shock and heavy depreciation of the domestic currency, you need to wait for a rise in nominal prices and expect a change in relative prices. That is what “exchange rate pass through” is about. And Turkey has it, although it has become less effective over the years with the growing sense of stability. But inertia still persists, as evidenced by the surge in the share of FX deposits in our banking system lately.

Thirdly, just look at the inflation rate in terms of domestic producer prices. The inflation rate was around 5.7 percent as of December 2015. In 2016, it had increased to 9.9 percent by the end of the year. This is around a 75 percent increase in the producer prices inflation rate, thanks to the depreciation of the lira. That leads to the rise in costs of production, together with the rise in the prices of imported intermediate goods. In 2017, we are also seeing a surge in the consumer price index to double digit levels.

Last year, Turkey changed the way it calculates it GDP. The timing was definitely wrong, though the change from a fixed base year method to a chain-linking of national accounts looks rather innocuous and legitimate. The EU has done it, and the whole developed world has been converted to this latter approach in GDP calculation. However, in countries of the “original sin” where economies are highly dollarized, chain-linking national accounts may not be better than the fixed base year method of national accounts calculation. Why?

In the chain-linking of national accounts, the change in relative prices is able to inflate growth rates. As a result, in a period where the change in relative prices is higher, as in 2016, one needs to be extra cautious in evaluating growth figures. That is why I think that growth in the first quarter of 2016 may actually be higher than the first quarter of 2017, even though the former was 4.5 percent percent and the latter was 5 percent. I advise caution regarding the job generation, wealth creation and debt repayment capacity of Turkey’s new economic growth figure.

My question remains: Is the chain-linking of national accounts an improvement over the fixed base year approach in a highly dollarized economy? I still do not have a definite answer. So is Turkey growing faster in 2017 when compared to 2016? I doubt it.

This commentary was published in Hürriyet Daily News on 24.06.2017