Last week, in correcting the statement in your article about Turkey that “per capita income in Turkey tripled in 10 years” (thanks for your correction in this week’s edition that in real local currency terms the increase in per capita income was only 43% during the time period in question), I referred to a term that I coined: ‘statistical urban legends’. I indicated that I define this term as ‘using numbers that are correct on the surface, to leave a misleading (that favors the government) and incorrect impression’. In the above article, the quote you have from the finance minister contains at least 3 examples of the practice of ‘statistical urban legends’:

1. By saying that “Public Debt/GDP ratio is just 36%” the minister implies that this is a low figure. This is correct if one uses developed countries as the benchmark. In fact, at the end of 2012, the GDP weighted average of IMF’s Fiscal Monitor ‘Advanced economies’ sample, which includes 30 countries, was 0.2.

However, comparing Turkey with developed countries violates the basic principle of the scientific methodology. The comparison needs to be made against countries that are similar to Turkey. The IMF publication mentioned above also has an ‘Emerging market economies’ sample (which also contains 30 countries), and the average debt ratio of this sample in 2012 was %35.2. Furthermore, the 10 country Emerging G-20 sample (which includes Turkey), had an average debt ratio of 33.7%.

I am not going to make a big deal out of Turkey actually having somewhat higher debt ratio than its peers. The point that I want to make is that the low debt ratio that the minister mentions is nothing special; other emerging market economies also have low public debt ratios. In fact, the same conclusion emerges if the other commonly used ‘fiscal discipline metric’ - Budget balance/GDP is used.

2. The minister is also incorrect when he states that Turkey has “low household and corporate debt”. I will not discuss in detail why the HH debt ratio of Turkey is not low. However, two points should be made: I. Again, comparing this ratio to developed countries HH debt ratio is not meaningful, on grounds that developed country HH’s have a culture of managing debt, while Turkish society is not yet at this stage and II. Consumer debt has been increasing at an incredible high rate of around 25% in Turkey during the last 7 years or so.

It is a well-known principle of corporate finance that the lower a firm’s business risk (unlevered beta), the more it is that such a firm can take on financial risk (debt). It is safe to say that the Turkish economy is more volatile than developed country economies. Additionally, the average (all-equity) Turkish firm is affected more dramatically from a given size shock than the average firm in a developed economy. These 2 facts argue that Turkish firms should have less debt on their balance sheets than their developed country counterparts.

The average Debt/Equity (D/E) ratio of U.S. firms is 50%. This actually overstates U.S. firms’ leverage ratio. Because, U.S. firms have record levels of excess cash, which should be thought of as negative debt. Average D/E of firms in EU countries is about 70%. According to an annual survey conducted by Istanbul Chamber of Industry, the average D/E for the largest 500 firms in Turkey was 120% in 2010, and increased to 141% in 2011. We don’t know the ratio for 2012, but my suspicion is that it increased further.

Additionally, of the $336.8 billion of FX denominated debt (2012) that Turkey has, $226 billion belongs to the private sector. Thus, in addition to having a high debt ratio, Turkish firms are also exposed to substantial foreign exchange rate risk. In fact, the significant decline in the value of TL since the recent protests against the government started, means that D/E in TL terms have increased even more

3. The minister’s statement that “the Central Bank (CB) has Large and rising reserves” represents another example of ‘statistical urban legends’. The government and the media frequently mention that “the reserves of the CB has hit another record”. What they ‘forget’ to mention is that the figures they cite is about ‘gross reserves’. Since the CB also has FX and gold denominated liabilities, what matters is ‘net reserves’. While gross reserves of the CB, as of January 25, 2013 was $123.3 billion, net reserves were only $29.170. On June 14 gross reserves was $128.8 billion (4.38% increase), net reserves was $29.7 billion (an increase of only 1.74%).

To put it differently, net reserves on the latter date was about 9 weeks’ worth of imports during the first 4 months of 2013. Additionally, given that Turkey has $156 billion of foreign currency denominated debt that will mature during the next 12 months (as of March 2013), the net reserves represent only 19% of this amount. This data clearly indicates that Turkey’s reserves are inadequate and the statement that Turkey has "Large and rising reserves” is incorrect.

Reply · Report Post