Black Sea puzzles

BEGINNING with the pamphleteers of the 16th century, several authors have pointed to the roots of Spanish economic woes and singled out the influx of Mexican and Peruvian precious metals, a phenomenon that is worth labelling Moctezuma’s curse. Moctezuma II – in some sources Montezuma II – was the Aztec emperor when conquistador Hernán Cortes landed in Mexico in 1519. The subsequent economic decline of Spain has endured for two centuries, and is known as an early instance of Dutch Disease. There is a vast literature that has examined since 1981 the ill-fated consequences on the manufacturing sector of the discovery of large quantities of tradable natural resources, a phenomenon known as the Dutch Disease after the impact of 1970s natural gas price increases on the Dutch economy. So far, the consensus view claims that Spain had fallen victim to an early instance of Dutch Disease.

Actually, there are no puzzles and Dutch Disease would be a far cry. Suppose the Black Sea deposit’s magnitude is confirmed. The natural gas deposit in question would be about 40% of what Romania extracts, and it is about four times as large as the Atlas field of Bulgaria. Also, to fix ideas, it is about 35%
of Leviathan and Tamar combined. It is a good find, but not so large as to be a game changer. Energy economists made a back of the envelop calculation that implies USD 2.5 to 4bn gas revenue per annum, which on average would cover about 1/10 of the current account deficit. Further finds could lead to other considerations. Also there is the recoverability issue, and there is the initial investment problem, and the time element. It would take at least five years to surface gas finds on average, even more. Well, could be helpful perhaps, but that’s about it. Or…


No, we should not. The Dutch disease phenomenon is generally associated with specific factors, the presence of learning-by-doing, rent-seeking behavior, human capital de-accumulation and suchlike. The Dutch disease is a disease only in so far as it discourages human capital accumulation, encourages rent-seeking, and leads to a loss of know-how and ability to produce in the tradable sector as a result of the discontinuity in the path-dependent learning-by-doing path, as Ken Arrow meant it back in 1962. All these have already happened here. It may possibly not get any worse than this simply because a mid-size deficit has been found. Yet you never know.


The most robust symptom of Dutch Disease is an overvalued exchange rate. Max Corden notes that in the presence of international capital mobility the exchange rate may appreciate more for a given rate of monetary contraction. Secondly, the exchange rate will respond more rapidly in this case both for expectational reasons and because capital movements are speedier than real expenditures. There may be overshooting as a result of these factors in the sense that the exchange rate could appreciate more than warranted and faster than expected. In other words, Dutch Disease might cause a bigger than expected current account deficit in the presence of international capital mobility. Furthermore, even temporary capital inflows can create additional demand for government securities and help reduce the real rate of interest. Falling interest rates can cause investment demand to increase. What Max Corden calls “exchange rate protection” may occur through direct intervention, sterilized through open market operations or by means of budget surplus, or both. This is the way to protect the tradeable goods sector. There remains of course the possibility that even if such intervention takes place the exchange rate can still appreciate. With an inefficiently high FX reserve carried by the central bank, the currency could still remain overvalued if inflows were to reach abnormally high magnitudes. In the end, Dutch Disease specifically entails a squeeze of tradeable production that is likely to become a long-term feature of the economy. We have been through all this before 2008. Now we have inordinately low Central Bank FX reserves, high corporate FX debt, a large short position, and a patently undervalued currency.


There may be other mechanisms though. Savings may fall because credit constraints are being relaxed as a result of capital inflows. In this case, if banks are bought by larger parent companies, i.e. European banks for instance or Chinese, and if non-core liabilities increase rather easily, then credit expansion is no longer dependent on domestic savings. The more banks rely on overseas funding, the easier for especially young generations to get access to credit. Assuming that the real interest rate is exogenous increases the model’s ability to explain the recent trend in the saving rates of different cohorts over the last decade. The largest decrease is observed for the youngest generations, which may be attributed to the relaxation of credit constraints, higher expected lifetime earnings and less uncertainty.

The table above is from Drelichman (2005) [Drelichman, Mauricio (2005), “The Curse of Moctezuma: American Silver and the Dutch Disease”, Explorations in Economic History 42, 349-380]. It shows in a nutshell the symptoms of Dutch disease as expressed by the behavior of key macroeconomic variables.

For instance, Coeurdacier et al (2015) find that a significant portion of the high saving rates for both old and young cohorts in China is due to the fact that China was more credit-constrained compared to the US at that time. In the less credit constrained economy, people spend more and save less. For Turkey, one may wish to consider the mechanism described by Altug & Firat (2018). One of the basic mechanisms that have allowed AKP to garner a voting stronghold in many regions is the credit expansion mechanism, a mechanism that has worked well between 2003-2008 and 2010-2018 twice. It seems to work today also, but there is limited scope left.


Two ideas emerge: First, a presupposition should obtain. There will be no “Dutch Disease”-type long-run repercussions after your domestic market is awash with financial flows. And any currency shock need not end up in a ‘sudden stop’, old school-type that is. If these obtain, then the second idea tells us that the exchange rate shouldn’t be misaligned for too prolonged a period. In other words, the lira shouldn’t be allowed to over-appreciate if foreign capital inflows anew. Typically, the “Dutch Disease” – ‘sudden stop’ type of phenomena occur after the initial financial liberalization. Add to this the lending boom that happened because banks are bought by larger overseas banks and may be allowed to get credit wholesale from their parent companies. Because of the ‘global savings glut’ or because major central banks cut rates
and eased quantitatively, there was cheap funding for many years.


Nevertheless, we have been through all this, right? In the first round, the lira over-appreciated. If sustained, this causes misallocation of resources and financial fragility. Now, the real exchange rate is the relative prices of tradeable goods in terms of non-tradeable goods. A real exchange rate appreciation – if it lasts for many years and if it borders over-appreciation – allocates scarce resources, mainly capital, to the more profitable non-tradeable goods sectors such as construction. These episodes need not be simultaneous. A t-1 episode appreciation that doesn’t extend to period ‘t’ can still cause a capital over-accumulation in the non-tradeable sector in period ‘t’ because production takes time. This is exactly what happened here after Lehman. Because myopia is very real and because the world was awash with cheap dollars again for a second round, capital accumulation basically took place in non-tradeables, especially construction, and manufacturing, the locus classicus of ‘catch up’ and productivity, lagged behind. The quality of GDP growth and its sustainability are being hampered as a consequence. Could USD 2.5 to 4bn per year change any of this in either direction?

Leave a Reply

Your email address will not be published.