TOPIC OF THE WEEK: Too hot to handle? Mapping overheating risks in Eurasia
Evaluating the medium-term government programs in Uzbekistan and Tajikistan has brought into sharper focus the very high growth rates in these economies, with neighboring Kyrgyzstan comfortably beating even those ambitious 7-8% plus growth projections. A question that I have posed for myself in recent months is to what extent these economies might be growing beyond their means. Heuristically, I have felt that Kyrgyzstan has been showing all the tell-tale signs of an overheating economy, while I have tended to give Uzbekistan the benefit of the doubt given the wide-ranging supply-side reforms it has been undertaking in recent years. Tajikistan has been much harder for me to pin down, so I have been agnostic in terms of possible overheating pressures.
In this report I have taken a purely statistical approach to addressing this issue in the CCA space. I use two different output-gap approaches—a more conventional quarterly HP filter and an annual production-function framework—to judge current performance relative to potential. They point to broadly the same country ranking. Kyrgyzstan and Georgia show the strongest excess-demand pressures, Armenia and Uzbekistan exhibit moderate ones, Tajikistan is more ambiguous, and Azerbaijan runs slightly below potential. The exact size of the estimated gaps differs across methods, but the broad cyclical diagnosis is relatively robust.
I augment this by modelling a logit/probit probability framework, which complements the HP-filter and production-function estimates by translating several cyclical indicators into a common cross-country score. This sharpens the cross-country ranking. Kyrgyzstan, for example, remains the strongest overheating case because both its output gaps and broader growth configuration point in the same direction. Georgia also shows substantial cyclical pressure, but the probability model places it below Kyrgyzstan. Armenia and Uzbekistan fall into a more moderate-risk category, while Tajikistan is downgraded despite very rapid headline growth because much of that growth appears consistent with a high underlying trend rate. Azerbaijan, by contrast, shows little evidence of overheating under either the output-gap estimates or the probability model. The results are usefully summarized in our table in the text.
Now read on...
Register to sample a report