Main /


Central Asia at a time of heightened uncertainty

29 April 2016
Authors: Клеменс Граф
Category: Other

Central Asia has been negatively affected by two shocks of a very different kind. Firstly, and probably at this stage more importantly, the sharp fall in oil prices and, secondly, the heightened tension between Ukraine and Russia and the associated disruptions to the flow of goods and capital. The two shocks are quite different in nature but both impose important policy questions.

The oil shock, in our view, is better thought of as a common shock across the CIS, to which the countries have reacted asymmetrically, and it is the asymmetric policy response rather than any fundamental asymmetry in the shock that gives rise to spillovers. The imposed restrictions on the flow of capital and goods between Russia and many Western countries in contrast only affect Central Asia indirectly, and here it is really the spillovers from Russia and Ukraine that matter.

I will mostly discuss the oil shock here, but clearly policy options will need to factor in the second shock as well. It is true that there are oil exporters and oil importers in Central Asia and hence there is an asymmetry in terms of the first impact. However, given the reliance on the flow of remittances from oil to non-oil exporters, essentially both groups face similar shocks to their current accounts.


According to the World Bank, remittances from Russia and Kazakhstan accounted for 38% of GDP in Tajikistan in 2012, 26% of GDP in Kyrgyzstan and close to 10% even in Uzbekistan, while oil exports accounted for 27% of GDP in Kazakhstan and 18% in Russia in 2014. Thus, with remittances being very procyclical, some of the oil-importing countries are arguably even more exposed to the oil price than the oil exporters. This suggests that the oil shock acts quite symmetrically across Russia and Central Asia as a large shock to the current account.


However, the response to the shock appears to have been quite asymmetric, possibly reflecting existing policy frameworks and the available buffers, but possibly also different views on the kind of oil shock experienced. Thus, while Russia has allowed the Ruble to float, Central Asia at large has so far opted to keep its exchange rates highly managed, and capital and current accounts restricted to varying degrees. It is this asymmetry in the response to the shock that leads to spillovers from one country to the other.


For policy makers, the choice of the response to the shock requires first of all an understanding of the kind of oil shock experienced. Despite the relatively similar dynamics of the oil price in 2014/15 to 2008/09, in our view the shocks are very different and hence potentially require quite different responses. The shock in 2008 was largely a demand shock, and while it was very large indeed -- and at the time it was not clear how long it would last -- it was still ultimately likely to be temporary. Instead, we think there can be little doubt that the shock in 2014 has largely been a supply shock, driven by a sharp rise in productivity in the shale sector in the US. This has essentially been recognised by OPEC, which decided not to target the oil price, at least until it has a better understanding of the cost structure in the shale sector. The decision by oil companies to structurally adjust their capex budgets also suggests that this is the accepted interpretation by industry insiders.

The nature of the shock has important policy implications:

  • Unlike a temporary shock, a permanent shock makes it impossible to rely on borrowing or existing savings alone to manage the adjustment, no matter how large the existing buffers might be. As long as the shock is permanent, any stock of buffers will be eventually exhausted.
  • There is a large increase in uncertainty and risk. Technological innovation essentially follows random processes, and hence it is rather difficult, or virtually impossible, to have a clear view of where oil prices will settle in the medium to long run. This is even more the case as the oil sector in many countries is so dominant that the shock to the oil price leads to a shock to the exchange rate; this, in turn, changes not only costs but also relative prices in other commodities, which in turn are important inputs into the production of oil. In short, relative prices between commodities and future exchange rates are in flux globally as seldom before, making it very difficult to estimate where future cost levels and exchange rates will ultimately settle.
  • It is not only the oil sector that is undergoing deep changes: the same is even more the case for the gas sector, a close substitute to oil, and hence the energy mix employed globally is likely to change, adding to the uncertainty on the demand side for oil. 
  • The new technologies employed in shale are by nature very different from the traditional technologies. Oil has always been a fixed-cost industry, while in shale production it is almost meaningless to distinguish between variable and fixed costs. The traditional cycle of 5-10 year lead times between the start of a project and the first barrel of oil being produced has shortened in some shale subsectors to below two weeks. Thus, if this proves true, we are unlikely to see a return to periods where oil prices overshoot because the supply in the short term is unable to react. In turn, this means that there will be no way for a swing producer to keep prices high enough to allow companies to earn their return on the fixed costs incurred. This, as well as the uncertainty about the “true” future cost level in shale, will make it very difficult to implement long-term, capital-intensive traditional oil projects.
  • Related to this, with relative prices that much in flux, there is a sharp change in the prospect of different kinds of oil projects. Productivity shocks ultimately lead to a Schumpeterian process of creative destruction. Thus, many higher-cost projects in deep water and oil sands, for instance, are likely to have to be put on ice for the foreseeable future.

All these factors have important policy implications for Central Asia and Russia. Oil reserves, or rather proven reserves (i.e., those that can be profitably extracted), will need to be reassessed and may turn out to be significantly lower than previously thought. This essentially lowers the wealth of many of the oil producers and requires adjustments to fiscal plans and future growth strategies.

Leaving aside these direct growth factors, a permanently lower oil price lowers oil export income and remittances, and changes the external balance. Under a temporary shock, in countries like Kazakhstan this could be dealt with by digging into the oil fund or borrowing in international markets. However, if the shock is permanent, this can only smooth the adjustment while ultimately a more meaningful adjustment through the adjustment of relative prices needs to follow. In this sense, the Central Asian countries and Russia all fortunately have their own exchange rates and currencies, and hence -- unlike countries in the Euro area -- can use them to adjust to the shock. Still, even if the exchange rate is allowed to act as the main shock absorber, there are many issues to consider and in a space as closely integrated as that between Russia, Central Asia and now also China, there are important spillovers that one country’s response poses for the choices of the other countries.

Given the asymmetry in sizes and dependencies, the likelihood of policy coordination is not very high. This is not new. The treasurer to US President Richard Nixon, John Connally, famously said “The dollar is our currency but your problem”. Likewise, the German Bundesbank arguably did not take into account the interests of other countries when setting policy post German unification, ultimately contributing to the breakdown of the ERM. Similarly, neither Russia, or for that matter China or Kazakhstan, are very likely to take into account the full effect of the spillovers of their decisions on others. In the first instance, this argues that the smaller countries need to wait for the larger countries to move first and react to the shocks before they can begin to make their choices.

Russia has opted for a structural response by allowing the Ruble to float, while China is only slowly introducing more flexibility into its currency and opening its capital account. The most important game changer, therefore, is the structural response by the CBR that Central Asia will respond to. Similar to the period post the Dollar float vs gold, Central Asian currency arrangements will mostly likely change as well. Currently most of the Central Asian countries have highly managed exchange rates and are restricting capital and sometimes current accounts to varying degrees. This was reasonably straightforward as long as Russia's currency was to some extent tracking Western currencies as well. With the Russian peg to the USD and Euro already a policy of the past, and China moving over time to floating as well, it seems highly questionable that the past pegs vs the USD or Euro that many countries in Central Asia were operating are still the optimal policy. Monetary policy frameworks will need to be fundamentally rethought. The options appear to be to a move towards a float or to re-peg against a different currency or a basket of currencies.

The choice between the two is a difficult one. Pegging to the USD had the advantage that Kazakhstan was effectively pegging to a country with a long history of a stable monetary set-up targeting low inflation, while at the same time providing smooth access to the most liquid international markets. Thus, while this still meant giving up independent monetary policy, at least Kazakhstan was well aware of the kind of monetary policy it was going to import. This is not to say that this was costless, as the credit bubble in the run-up to 2008 showed, when Kazakh banks were able to borrow at foreign interest rates that were too low for Kazakhstan, while the banks and borrowers, potentially due to the peg, were happy to take on too much foreign exchange risk.

Any re-peg would most likely need to tie the Central Asian currencies quite closely to the Russian Ruble. While this choice may not appear that obvious from just considering the relative weight of the trading partners, it is the common exposure to the oil shocks that would make this choice quite natural. However, while the CBR is well on its way to establishing a credible framework to switch from an external monetary anchor to an internal one, inflation, this is necessarily still very much work in progress.  Monetary arrangements in China are even more in flux. Thus, the typical benefit of importing credibility by pegging is not as obvious. Also, neither Russia nor China have deep financial markets that would be immediately beneficial for Kazakh borrowers to tap into. Lastly, as we said above, Russia is currently also experiencing an idiosyncratic shock which restricts its access to international capital and has put pressure on the Ruble. The CBR will also need to take that shock into account in policy making, which could make its choices less obviously beneficial for Central Asia; hence, Central Asian countries will need to bear this in mind when deciding whether they want to opt for a peg to the Ruble.

Apart from the choice of what to peg to, there is also the question of the level to consider, which brings me to the second characteristic of the oil price shock: large uncertainty about the cost level of oil projects. Traditionally, markets have viewed Kazakhstan’s real exchange rate vs the Ruble as the anchor. However, currently this is not so clear, if it ever was. Apart from the sanction shock that Russia is experiencing, the shock to oil prices is also non trivial. Lifting costs in Kazakhstan differ from those in Russia, as does the state of the investment cycle, and hence the impact that the oil price shock has on the two industries may not be symmetric. The country that has the higher lifting costs could potentially need to depreciate further than the other one. Thus, any chosen level of a peg may need to be reconsidered depending on where oil prices finally settle.

Given the uncertainties involved, some Central Asian countries may well choose to opt for a more flexible exchange rate and instead rely on inflation as an anchor, at least until the level of uncertainty is reduced around both the sanctions and the future level of oil prices. Post the breakdown of the original European Exchange Rate Mechanism (ERM), countries first opted to widen currency bands very significantly from 2.25% to 15% before some eventually opted to tie their currencies fully to the Euro. These wider margins freed the hands of many central banks in reacting to the shocks that had put pressure on the original arrangement.

Similarly, many Central Asian countries may opt for more flexibility to allow their exchange rates to adjust as the oil price shock plays out, as well as waiting to see what happens with the restrictions on capital and goods flows in Russia. Such a response clearly puts the emphasis on building on and investing in the credibility of the region’s central banks in managing inflation and financial stability. Arguably, reducing the level of NPLs and increasing transparency in the banking sector are factors that in any case should be addressed prior to tying monetary policy fully to an external anchor.

It is understandable that many central banks in Central Asia are keeping their options open in this period of high uncertainty. However, there is clearly also a cost in sticking to arrangements that are unlikely to be sustained in the longer run, such as an increasing level of dollarization and risks that hamper financial sector development. Given the difficulty involved in eliminating those risks completely without tying one's hands, monetary authorities may consider reducing the level of risk through other means. A choice of currency regime may not be the only way to address the uncertainty faced by Central Asian economies; policymakers could also consider hedging some of the original risk stemming from the oil price, given that such operations do not pre-commit and have the potential to reduce risks to the current accounts in the current environment. 

You have to login to post comments.