In Part I of this article, we recalled some basic economic theory related to the exchange rates and reviewed some of the downsides of real exchange rate appreciation. Also, following question 1, we concluded that the net effect of exchange rate appreciation on the economy today and growth prospects in the future are likely to be negative, and largely so, for an economy like Armenia's.

Let us now turn to question 2 and discuss the conditions under which a policy response would have been warranted, and what type of policies could have been successful in curbing the exchange rate appreciation and reducing its impact. I argue that the overall response of Armenia's policymakers to these persistent capital inflows was grossly inadequate, as they failed to apply any counter policy measures, including those to be discussed below. As before, references to economic theory and some common sense will be inevitable here!

These same economic theory and common sense tell us that it would be sensible to intervene and try to reverse exchange rate appreciation if: (1) the appreciation is likely to result in irreversible consequences for the economy, in terms of the damage to its production and export capacity, and (2) the underlying factors behind the appreciation are temporary or at least will have a declining intensity in the future. And you probably thought this is going to be straightforward! But while these are not easy issues to tackle, we should nevertheless try to discuss them to find out whether intervention would have been warranted.

I already eluded in the first part of the article to the fact that the workings of the productive sector of the economy depend to a large extent on the degree of financial sector development. Sophisticated financial markets allow companies to borrow more to weather various shocks. In a developed financial system, banks also require less collateral against loans they extend to companies, and typically can issue them against the companies' future cash flows and not necessarily against (hefty) collateral. So when a company faces difficulties in an economy with developed financial system and has to cut back production or close, this would not have a major impact on its production and employment. Even if it does, the impact is likely to be temporary because companies can borrow to stay afloat and to invest in new technologies, if they have to, and recover. In contrast with this, in an undeveloped financial markets' setting, bank credit is scarce and therefore could not act as a buffer against the shocks that hit companies. Therefore, the impact on the productive sector is more severe than it otherwise could be. Here is where Armenia is on this scale—its financial markets are underdeveloped (if my memory serves me right, ranked low even by regional standards) and collateral-based lending, unfortunately, is still the predominant way for banks to deal with companies. Therefore, when an exchange rate appreciation (or any other shock) of the magnitude that took place in Armenia hits the productive sector, rest assured that it will have an effect on the producers that will be deep and won't go away soon, if it ever does! Part of the reason why the impact will be severe is also because Armenian exporters are already operating with very thin margins given the exorbitant transportation costs and corruption in customs. So judging by criteria 1 above, policy intervention—either on the exchange rate side or elsewhere, through perhaps government-administered loan programs—would have been necessary to prevent appreciation from having an (irreversible) impact on the economy.

Regarding the resiliency of capital inflows, which are the underlying reasons for the appreciation (criteria 2 above), we need to answer two questions: (1) are these flows completely independent of the factors in Armenia (e.g., policy measures, economic fundamentals, etc.) and (2) are they going to remain at the same level in nominal terms (and thus go down relative to economic activity in Armenia), or are they going to grow along with the economic activity thus continue increasing in nominal terms)? While the discussion of these issues is much too technical and goes beyond the scope of an article like this, it is my firm belief that these flows are both dependent on the factors (including policy measures) in Armenia and likely to decline as a share of overall economic activity (e.g., GDP) in Armenia. (I will be happy to discuss my thoughts on this separately, which are based on foundations of modern international economics as well as the underlying demographics of Armenia's remittances). Thus, this criterion too suggests that a policy intervention would have been desirable.

How to Intervene?

Having established that policy interventions to prevent the foreign currency flows from appreciating the exchange rate would have been warranted, let us look into various policy choices that were available to Armenian policymakers, but were not used or used with insufficient vigor and commitment. I will group these policy choices between those available/attributable to the Central Bank (CBA), the Ministry of Finance (MOF), and the State Anti-Monopoly Agency.

Toothless Central Bank, or Too Little, Too Late

Policy Measure 1: “Intervene and sterilize” to reduce the pressure on the exchange rate but also to keep the money supply unchanged to avoid inflationary pressures. This, however, requires a portfolio of government (or Central Bank-issued) paper/debt, which the Central Bank did not have. (We will discuss this issue further, in the context of the Ministry of Finance's policy measures). While the CBI should not be fully blamed for this, what it could have done is to (i) promote the public's understanding of the issue, so the latter could place the blame properly, with subsequent pressure to change the course, and (ii) issue its own paper/obligations to use for sterilization. While the Central Bank did start issuing its own obligations and selling them on the market, this was a drop in the bucket and was done pretty late in the day, and therefore was largely ineffective. (To help you digest this economic jargon, here are some explanations. An “intervention” is a purchase by the Central Bank of foreign exchange, which has an effect of increasing the amount of money in circulation, otherwise called the money supply. “Sterilization” is referred to as a sale of treasury bills (T-bills) and other government/Central Bank obligations by the Central Bank with an aim of reducing the amount of drams in circulation. So while an “intervention” by the Central Bank to buy dollars increases the amount of drams in circulation, a “sterilization” operation could bring the amount of drams in circulation back to the same level before the intervention. Of course, the money supply is a key determinant of inflation and, therefore, needs to be watched carefully).

Policy Measure 2: Signal a more aggressive policy stance that would indicate that the Central Bank is committed to maintaining a “socially optimal level of the real exchange rate” along with its main policy target of inflation, no matter how vaguely this term is defined! As many things in life (!), the CBA's ability to conduct exchange rate policy depends on how its signals its (policy) intentions. This is a very serious game that could end up putting Central Bank on the losing side if it fails to play properly. One way to win this signaling game—that is, of course, if you are serious about it—is to announce that you stand ready to smooth any unwanted short term fluctuations or odd (i.e., not driven by economic fundamentals) market sentiments, which is so common and potentially so destructive in today's world. Intervening in the right way (i.e., selling foreign currency when you have to sell it and buying when you have to buy it, and not the other way around) usually helps too! The CBA, however, had lost this signaling game when—in addition to the demoralizing messages coming from its Board—it allowed the Ministry of Finance to publish its “assumption” of the average exchange rate as part of the 2007 budget presentation. And guess where the exchange rate was placed in 2007 budget—at mid-300s per dollar, indicating a further appreciation of the dram relative to 2006 level….as if the economy didn't have enough of it yet! The logic of revealing the exchange rate assumption embedded in the budget in the context of a country like Armenia (which is undergoing a socially and economically painful process due to exchange rate appreciation, and where economic agents—both the exporters and those who receive remittances in foreign exchange—were longing for a different message) completely escapes me. Armenia's budget has no major sources of foreign exchange-denominated earnings, nor it has any major foreign exchange-denominated expenditures for the MOF to want to be “open and transparent” about its macroeconomic assumptions. The exchange rate issue is not even in the Ministry's policy domain, so why let them reveal the exchange rate assumption? Signaling, I guess….but the wrong type!

“Detached” Ministry of Finance, or Why Should We Care?

Policy Measure 3: Supply the Central Bank with enough T-bills for the latter to use them to sterilize after intervening. Yet, the CBA was not granted access to sufficient amounts of T-Bills to do this, which led to the second best option chosen by the Central Bank, that is, the “do-nothing” response. (Faced by a shortage of T-Bills, the CBA cannot intervene, because it will increase the money supply and inflation, an undesirable outcome given that inflation is the primary target of the CBA. Contrary to common belief, CBA is not charged with stabilizing the exchange rate—it is only charged with stabilizing prices). Without the MOF's help, the CBA with its small and already heavily dollarized balance sheet could do little in this respect. Critics of this approach have implicitly argued that the increase in interest rate as a result of issuing more T-bills will hurt the real economy (and those same exporters) through higher borrowing costs. My response to this is that this is unlikely to be an issue. There are plenty of liquidity in the market and not enough secure projects for the banks to finance, so they will be happy to hold T-bills without charging higher interest rates for that. Even if the yield on the T-bills goes up a notch, the true borrowing costs for companies are much higher and are unlikely to shift upwards immediately, if the banks behave in a competitive fashion. In addition, critics argued that sterilization is costly for the budget and, therefore, should not be undertaken. Sterilization indeed has costs measured both in terms of the interest to be paid by the budget on the additional T-bills it will provide to the Central Bank, but also in terms the interest rate differential between the T-bills and the dollar-denominated reserve assets that the Central Bank will have to hold after intervening and sterilizing. However, the cost sharing with the rest of the economy would have proved a good policy response, given the downside and the social costs of the alternative (“do-nothing”) policy.

Policy Measure 4: Provide carefully targeted subsidized loans to exporters to prevent irreversible impact on their productive capacity. This could have been done in a way to encourage some of the most vulnerable exporters to switch to more efficient technologies using the relatively inexpensive funds made available through the budget (via a commercial bank or a specialized government agency). Critics would say that subsidizing is not “kosher”, and therefore should not be undertaken. My response would be—think about the alternative and get over it! A properly designed small-scale program to assist most vulnerable exporters would have generated high sizable social benefits without jeopardizing fiscal stability.

Policy Measure 5: Increase expenditure on imports of foreign investment goods and by doing so reduce the pressure on the dram (by demanding more dollars). In addition to expanding the overall fiscal envelop, the Ministry of Finance could also conceivably shift the composition of fiscal expenditures from domestic to foreign. This could ease the pressure on the exchange rate market but also keep domestic prices low (by reducing the demand for domestically produced goods and services). Every economist in his/her right mind knows that Armenia's fiscal spending and overall deficit do not match its status of a developing country which is trying to build long-run growth capacity. Therefore, additional spending on imported investment goods would not have hurt the fiscal sustainability and would have helped strengthen the supply side and better position it for growth in the future.

Policy Measure 6: Tax the “consumption” of real estate. While imposing capital controls to prevent the dram from further appreciating would have been costly for Armenia (given its desire to attract capital) and ineffective (given the channels through which these flows are transferred), changing the destination of those flows away from the construction (and consumption!), would be the way to proceed to avoid Dutch disease-type adverse effects on other sectors of the economy. Construction sector is hardly taxed and is partly the reasons behind Armenia's miserable tax collection record. Not taxing construction results in over-expansion of that particular sector at the expense of other sectors, since construction competes for inputs of production—labor and capital—with other sectors in the economy. Taxing construction is also likely to put the sector on a path of a more stable growth and work as insurance against bursting of the real estate bubble in the future. (You essentially would pay more now as a result of higher taxes but avoid severe price drops in the future, since the sector will grow more moderately this way). Critics of this approach argue that capital/remittances flows are elastic and would go down if the policy starts targeting the reason why they are there (implying that the money comes from the Diaspora Armenians to largely buy real estate in Armenia). This could well be the case. However, given the inelastic nature of Diaspora Armenians' sentiment towards Armenia (Here, I said that!), taxing construction sector is unlikely to reduce the total flow of transfers and other Diaspora-driven flows. It will instead help re-direct those flows elsewhere, to other sectors of the economy.

Useless State Anti-Monopoly Agency, or We Aren't Paid Enough to Deal with the Oligarchs!

Policy Measure 7: De-monopolize import sector to allow the price and exchange rate signals to be fully and quickly transmitted to economic agents (i.e., producers and consumers) and policymaking institutions (e.g., the Central Bank). Let us recall the discussion we had in Part I. If the prices of imported goods in dram equivalent were allowed to adjust fully to reflect appreciated exchange rate, this would have resulted in lower overall/average prices. Since this would imply deflation, it would have allowed the Central Bank to intervene on the exchange rate market without the need to sterilize the results of the intervention! (Think of it this way: because the domestic prices of imported goods dropped, the Central Bank has now more room between its target inflation and current (lower) price level. Therefore, it can “afford” to buy more dollars in the market to help prevent the exchange rate from appreciation). In this case, no additional T-bills would have been required and there would have been no issue with the interest rate possibly going up in the case of sterilized interventions. Critics of this measure cite the political-economy constraints faced by the authorities and, therefore, the high “social” cost of doing something in this regard. It turns out some importers in Armenia are characterized as having strong government links (if not operating under the protection of the highest layers of the government) and are seen as increasingly difficult to deal with. I will decline to respond to this issue, which is largely outside of the scope of this article and has to do with the political will to change things on the ground and has very little to do with economic policy.

Winners and losers

In closing, I would emphasize again that policymakers in Armenia had at least these 7 and possibly other measures to reduce the impact of appreciation on the economy. Of course, the nice thing about having a few measures at your disposal to tackle an issue is that you do not have to go all the way with any of one measure. Instead, you can rely on a carefully designed and orchestrated combination of these measures to be undertaken simultaneously to address the issue—they may also end up reinforcing each other that way. Alas, it my view, the response was grossly inadequate and, unfortunately, we will be seeing the fruits of this for months (if not years) to come.

So all in all, what's really at stake here and who are the winners and losers of this?

Losers

•  Exporters (through loss of competitiveness) and the share of population whose livelihood depends on the export sector (through loss of jobs and cuts in salaries);

•  Families who receive remittances in foreign currency (through the decline of the purchasing power of their remittances);

•  The Central Bank (through the loss of credibility).

Winners

•  Importers.

While I would not join those who argue that the exchange rate has been manipulated intentionally to benefit a few powerful importers (perhaps a punishable offense by itself), I do think that the fact that policymakers failed to adequately address the underlying problem and its consequences is almost equally bad. After all, the outcome is still the same—importers got richer at the expense of pretty much the rest of the economy and the policymakers could have prevented this from taking place. Who is to be blamed for this? I am afraid, this and many similar questions are likely to remain unanswered in today's Armenia.

Njdeh Melkonian is an economist based in Indiana, United States, who specializes in economic and developmental issues, including those facing the economies of transition.

March 26, 2007

Njdeh Melkonian