The saga of the business cycle, Turkish style
After the 2012 Article 4 of the IMF, which purported to issue a stark warning to Turkey, many a smart cookie began thinking the 2001 Kemal Dervis-IMF program had completely died off. The staff paper accompanying Article 4 was even more concise; but it was also bold. Accordingly, there would be no way such an economy could be sustained going forward, given that the IMF had predicted a financing need of $317 billion in 2017, and of course a massive current account deficit (CAD). Because the CAD had edged up to c. 10% of GDP in 2013, and because nobody had predicted a drastic fall in oil prices at that time, 2013 and 2014 would have been years of stagnation, autumn leaves would have fallen, and after 2015 the gateway to hell would have been wide open. Did that nearly apocalyptic vision come true? No, but it embodied a germ of truth nonetheless. At the very least there exist cycles, and Turkey could well approach the through of a cycle back then. It didn’t exactly happen due to many factors – cross-border capital inflows, the buoyancy that lasted more than envisioned in the construction sector and suchlike, but it is a cycle after all, isn’t it? Are those cycles concomitant with the global – at least non-local – cycles, or are they mainly homemade? Both, I think. Thus, the chronology of local cycles is the main theme today.
Cycles for all seasons
There are varieties of business cycles. There are Kondratieff cycles, medium-term cycles, short-term cycles and so on: they may be superimposed at times. The graph depicts important post-1960 turning points. 1965 is the first peak year, but momentum is immediately lost. Momentum of the same magnitude is observed c. 1973 due to sharply rising workers’ remittances. 1973-77 is an interesting period: the consequences of the oil shock were not translated into domestic prices, and the Cyprus Operation took place. The US embargo took off in 1975. 197780: the slippery slope towards the economic decisions of Jan. 24 and the coup d’état. The economy dived down and rebounded after Prime Minister Turgut Ozal. This momentum ended in 1989, and the capital account was opened up. The second through, though not as deep as 1980, occurred in 2001. As the Justice and Development Party (AK Party) decided to implement the Dervis-IMF program in 2003, the economy jumpstarted in a way reminiscent of Ozal’s heyday. The AK Party momentum dies off in 2006. After Lehman enters the third through. The amplitude is not as large as the previous two crises, but it is every bit as real. After 2008, per capita income flattens. Hence, the following analysis. (1) The 1965 momentum was due to import substitution and the first Five Year Indicative Plan, but did not last long enough. (2) Switching from import substitution to outward looking policy was debated in 1971, but no decision won out. Instead, in lieu of exports workers remittances entered the picture in 1972. The 1973 momentum was “manna from heaven.” (3) 1980, the beginning of Ozal’s reforms, the whole world outlook changes, and so does the economic spirit. The impetus lasted nine years. (4) 2001: end of another era. Dervis steps in and opens up five years of momentum. (5) 2008: end of the Dervis program. The Turkish economy had already lost stream before Lehman. (6) No new program, no renewed momentum. Per capita income stagnates. (7) The new GDP series changes the picture, but it poses new questions and creates more puzzles than it solves.
On the other hand, if we look at the world per capita series depicted in the graph, we would observe that they are barely concomitant with the Turkish cycles. Even the adverse impact of the 1973 First Oil Shock could have been postponed here due to remittances, and it did only translate into inflation and other economic calamities after 1977. World cycles, at least in the past when the economy was still a close economy, played a role only as constraints perhaps, and their consequences were felt with considerable leads and lags.
One can only hope that the incredible qualitative changes the new GDP series suggest can only be remedied as time goes by because the chain index is automatically updated annually. The past data loses its significance. We can’t exactly calculate the contributions of domestic and external demand as sources of growth for instance, but this is not that important. Calculations of fixed capital formation did go too far, and although real estate plays a more important role – as it deserves – in the new series as capital, the implication that there have been more than enough savings is not even ridiculous. Because the CAD is a datum we can compute, as fixed capital formation is jump-started, then the savings component that has to have financed investments skyrocket, and the Turkey of the last seven years looks like China. I don’t buy this kind of overzealous argument on account of a number of correlations that don’t square at all, although I accept that construction & real estate development are also “capital” of a kind.
A d fferent approach
Now consider another approach to the local business cycle, one that has recently been put forward by Dr. Osman Sari in a working paper (“Turkey’s Economic Cycles”, Politikekonomi.net WP 2017-1). The main take away is twofold. Firstly, the cycle is local rather than globally induced, and it is not
periodic. Secondly, there are periods extending roughly over two decades that characterize the cycle. If we hold this second conjecture true, the cycle is like a 15-20 years Kuznets swing. If we consider the failed Tablita of 200001 as the through of the previous cycle, then the current cycle is almost over. Of course, it is difficult to consider the 1940-60 period as one cycle because the WWII is almost a cycle in itself and because the lowest world GDP annual growth rates were recorded not only during the 1929 Great Depression, but also during WWI and WWII, and in their immediate aftermaths. However, the periods 1960-80, 1980-2000, and the post-2000 period, although they can be divided into sub-periods, are obviously clear-cut devices because they correspond to largely homemade, macroeconomic sea-change induced cycles. They all began with an aura of success, the finally found cornerstone of stability and prosperity, and were hailed towards their mid-lives with laurels that bordered euphoria. Then, the victorious march turned into slouching towards the drift, as one graphic of Dr. Sari documents.
Cycles are not periodic. Cycles are not replicas of each other. There are booms and busts, peaks and throughs, but neither the duration nor the skewness nor the kurtosis need be the same. Twenty-year episodes may or may not be telling. However, any stimulus provided by a sea change lasts c. 5 years and then comes stagnation. Dragging one’s feet doesn’t help one to jump-start, and move ahead with flying colors. When momentum is lost, fine-tuning with either reforms or with enhanced financing capacity doesn’t secure the long-run. Cycles can be said to be endogenous in the sense that the policy changes that induce them are captive in the dire needs the old model has led them through: macroeconomic vision- ary orientation changes have become necessities. This is why radical change is implemented in the first place.
Early years of cycles always br ng success
Nevertheless, once implemented the new macroeconomic orientation not only slowly feeds into fundamentals, but also starts a new cycle by providing unknown stimuli. The first years of the new strategy are always a success. It was a success after the 1950 elections because agricultural (export) prices were very high in the early 1950s, and because USD flows were secured through the Marshall Plan. It didn’t last into the late 1950s, and balance of payments deteriorated, ending in the 1958 devaluation. The first indicative five-year plan of the early 1960s was also initially a success. Import substitution changed the whole playing field, and could be extended for a few years more in the 1970s thanks to unexpected workers’ remittances pouring in from Germany. This episode also ended badly, and the 1980 program levelled the playing field once more. This time round the new neo-liberal vision was dubbed outward-looking policy and export-oriented strategy. Turkey’s strategy of reorienting its economy towards export-led growth resulted in a substantial increase in the value and volume of exports during the 1980s. Increased exports of manufactured products constituted the major source of growth in goods exported. Increased revenue from tourism and other services generating foreign exchange also contributed to an improved current account. Hence, it worked initially until such time as, under the pressure of political competition, public spending skyrocketed, and the capital account was opened up in 1989 earlier than it should have been. Hence, it was hoped that foreign investors would buy Turkish domestic debt. The roller-coaster type of zigzagging in the 1990s resulted in high and chronic inflation, a fragile currency and very high real interest rates. In fact, as we approached the 2001 crisis, the political economy of Turkey could be characterized by two concepts: incentives and taxes. Throughout the 1990s, Turkey was journeying through the vicissitudes of a single concept, one entwined with many of its self-generated economic conundrums: taxation –or lack of it. Coupled with high public spending constantly triggered by upcoming electoral cycles, this could only mean one thing: a huge public deficit and a huge public sector borrowing requirement. The failure of the Tablita of 2000-01 warranted another program, and the Dervis-IMF joint effort came to the rescue in 2001. Again, this program also worked. The first phase was a clear success, until 2006-07. The household debt graphic depicts one facet of the post-2001 cycle, the credit boom of 2004-05 and the accompanying enormous increase in household debt. This graphic (discontinued series) has had the potential to show how household debt could tend to swallow up to two-thirds of disposable income. Indeed, it is becoming increasingly costly and difficult to keep consumer demand going, despite the fact that in Turkey people do tend to consume always.
Conclusion: Turkey induces its own cycles basically, and imports economic phenomena from overseas only with delays. After 2001, the economy has become increasingly dependent on capital (portfolio) inflows, but the capacity to replenish the latent dynamics lies with the administration. A new macroeconomic strategy, a new cycle is what most economic actors have been waiting for since at least 2012, and no matter what may come, history shows the first years of any ‘new’ program always brings success.