I appreciate the opportunity to clarify some points raised in the recent discussion in this forum and also respond to the question posed by Mr. Gabrielyan in his follow-up remarks.

First of all, one should be careful not to mix and match items from the Current Account (CA) and Capital Account (KA). It should be noted that in the absence of reserve accumulation by the Central Bank, the current account equals the capital account. In other words, when CA is in deficit, the KA will have to be in surplus by exact same amount to finance the deficit in the CA, and the other way around. While items such as (i) trade in goods and services, (ii) official and non-official transfers, and (iii) factor income belong in the CA, items like (i) foreign direct investment and (ii) portfolio investment belong in the KA. We cannot just pick the ones that suit the point we are trying to make and say that there is a lot of (or not enough) pressure on the currency. We need to look at either CA or KA, not both, as the other one by definition would be the exact mirror imagine of the first, assuming no reserve accumulation by the Central Bank. Let us agree to focus on the CA, because most readers can relate to it easily, as it contains imports and exports of goods and services, remittances, and compensation of workers working abroad. [1/] Indeed, when one looks at Armenia’s consistently negative CA the story is clear and unambiguous-the dram should have lost value and not gained value, due to the fact that the demand for drams (i.e., exports and all other credit items in the CA) is less than the demand for dollars (i.e., imports of goods and services and all other debit items in the CA). [2/]
 
My second point relates to the predictability of remittance flows both in nominal terms but more importantly, compared to the size of the economy, measured by GDP (question posed by Mr. Gabrielyan). As I previously eluded to in my main article, this is very much a function of the demographics of the remitters and also the developments in both host and home countries of the remitters. A casual study of empirical literature on remittances world-wide would reveal that transfers to home countries typically resemble an inverted U-shape relationship as a function of time away from home country, indicating that transfers initially go up (as the remitters settle in host countries and start earning more money) but then come down (as ties with the home country decline, and family members, whom the transfers were intended for, die. The pressure to send money home also declines as the home country starts doing better economically, and the need to send money home is not as strong). [3/] To complete the picture of remittances flows in the Armenian context, one needs to put this behavior on top of the fact that the migration from Armenia has virtually stopped in the past couple/three years, suggesting that the total number of people who would be sending money home has probably reached its maximum already. When one puts these two factors together, the picture that emerges is one where remittances flows will have to slow down and then decline at some point, no matter how rich the remitters become (including because of higher oil prices in Russia). Now, when you measure these remittance flows relative o Armenia’s GDP (which in recent years has grown considerably, especially in US dollar equivalent) the picture would be even more pronounced-the ratio of remittances to GDP would decline even faster, because GDP grows at a quite high rate. Figures on total remittances to Armenia presented by Mr. Gabrielyan would certainly support this declining trend as a ratio to GDP. If so, they would also support the fact that remittances declined relative to the size of the Central Bank or Armenia’s monetary aggregates (because they both grew faster than the GDP in recent years), which is what really matters if one were to judge whether the Central Bank can deal with those flows properly or not. It is true that predicting this decline in remittances requires some understanding of the underlying processes, but it was not impossible. The only uncertain aspect in this is how long it will take for the transfers to reach their peak before they start slowing down. But estimating this should not be a problem empirically if one is armed with proper tools and desire to “get to the bottom of it”.
 
Having said this, once again I would like to emphasize that I do not think that the blame for the poor handling of the pressures on dram lies only with the Central Bank. I think the Ministry of Finance, Ministry of Industry and Economic Development, and the Anti-Monopoly Committee are equally guilty, if not more so. The reason is because they failed to formulate a correct economic development vision and to understand how Armenia’s economy should function if it wants to succeed and compete internationally. Armenia, a land-locked country with hostile neighbors and huge transportation costs (last time I checked it did not have a direct sea access!), should not have had to go through a 40 or so percent real appreciation in 2 years. If the policymakers in those ministries have given the concept of Armenia’s growth any thought at all, they would have probably come to a conclusion that Armenia’s progress has no choice but to be export driven, since its own internal market is small and in most cases would not allow companies to consolidate to benefit from economies of scale and to innovate via research and development, both necessary factors for growth.

Unless companies consider Armenia a place where they can produce and export from, very few of the big and growing ones are likely to do business in Armenia. And then the country will be stuck with a construction-sector-driven growth, which has the word disaster written all over it (notwithstanding the fact that as many lucrative sectors in Armenia this one too is concentrated in the hands of a few government-connected “businessmen”).  No matter what the Central Bank or other observers say regarding the impact the appreciation had on Armenia’s exports in the recent year or so (and of course they claim that the negative impact was mild), one thing is absolutely incontestable-NOBODY really knows what the export growth would have been had it not been for this huge and rapid appreciation. What makes economics a fuzzy science is that it does not allow for what they call “counterfactual experiments”. In other words, you cannot repeat the same experiment by keeping everything else the same and just change one factor that influences the outcome (in this case the real exchange rate appreciation). Economists can only GUESS the outcome, knowing what the relationship is likely to be between that factor and the outcome. And when one takes into account that exporters could have had 40 percent more revenue in their hands if it were not for this appreciation than what they have now, one can only GUESS that the result would have been much more favorable for them and for the economy as a whole. Most countries in the region and Armenia’s main comparators have undergone similar pressures on their currencies, some from the CA (via remittances, like Moldova) and others through KA (via oil sector investments, like Azerbaijan). Yet, to the best of my knowledge none of them experienced appreciations of this magnitude, because the policymakers in those countries most likely understood that they need to step in and share the costs of rapid appreciation with the exporters to avoid deterioration of their countries’ export potential. Armenian government should have done the same-it could have slowed down the appreciation and allowed the export sector to adjust to the shock of appreciation. For some reason they decided to hit the exporters with the bill upfront, perhaps thinking “why prolong the agony if they could force the patient go through  shock therapy now?” Well, I think they got it wrong, as they may have killed the patient who did not quite enjoy the shock.
 
Footnotes:
 
1/ By proposing to concentrate on the CA, I am not suggesting that KA-related flows cannot be critical in determining the exchange rate. I simply choose to focus on CA (and not KA) because the flows in KA are not as important in the Armenian context and do not by themselves pose any major risks in terms of pressure on the currency--the gross flows in the Armenia’s CA are much larger than those in the KA.
 
2/ It is true that he size of the CA deficit would not tell you by how much the currency would depreciate, since that would depend on where the CA stands vis-а-vis its equilibrium level in the medium term. But it would nevertheless tell you which way would the pressure on the currency go in the short run. And when you have a CA deficit, it is likely that the pressure in the short run is in the direction of depreciation and not appreciation.
 
3/ The picture would not change much if one factors in the presence of seasonal workers, since this component too has features similar to that of permanent migration, with similar implications for remittance behavior.

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

Njdeh Melkonian