J-walking

Why do we adjust to new exchange rates so slowly?


In 2008, a real estate bubble burst in Kazakhstan. Foreigners yanked their dollars out of the banks, fearing that they would collapse (and, with foreigners’ help, they did). Because they anticipated (incorrectly) that an ensuing recession would be more severe in Kazakhstan than in the United States, people sold their tenge to Kazakhstan’s central bank in exchange for dollars. Official reserves of dollars emptied.

Without dollars, the National Bank of Kazakhstan would not have been able to defend the tenge against financial shark attacks. And so, in February 2009, the National Bank devalued the tenge, from 120 tenge per dollar to 150 tenge. The Bank hoped that by increasing the tenge cost of buying a dollar, the devaluation would stop the runs for bucks.

The Bank guessed right. But its 25% devaluation sparked controversy for two reasons. First: It cut the value of the banks’ tenge-denominated assets by a fourth, increasing their chances of bankruptcy. Commercial bankers got a Mohawk haircut.

Second: It cut the value (in dollars) of the nation’s sales of goods to foreigners, net of what it bought from them. Before February 2009, Kazakhstan could sell a barrel of oil for, say, 12,000 tenge, or $100. After the devaluation, that barrel was worth only $80.

This problem is general. A devaluation of the tenge reduces the amount of foreign goods that it can buy. A given amount of Kazakhstani goods sold will purchase fewer foreign ones than before. The value of our net sales will fall.

Clinging to contracts

For example, suppose that we trade to Europe 100 barrels of oil for one automobile. Also suppose that we sell in all 1,000 barrels of oil in exchange for 10 autos. After our devaluation, the terms of trade may become, say, 150 barrels for an auto. We thus pay 500 more barrels than before for 10 autos. Our net sales – the “balance of trade” -- fall. Over time, Europeans will realize that the price of our oil has fallen, from 1/100 of an automobile to 1/150, and so they will buy more barrels. This will improve our trade balance. Over time, it will resemble a J – declining, then rising.

The J curve attracted the attention of economists in the early 1970s, when the fall of the dollar failed to increase quickly the value of exports, net of imports, for the United States. Stephen Magee suggested a three-period analysis. In the first period, the old trade contracts are still in force, so the devaluation affects neither prices nor quantities. In the second period, prices adjust to the new exchange rate, but quantities are still fixed. In the third period, quantities finally react to the change in price.

If we devalue our currency, then our balance of trade will improve for sure in the third period, when our lowered prices will increase demand for our exports. Prolonging the first two stages may delay this improvement. That may have happened to Kazakhstan in 2009, when uncertainty about the world economy may have induced oil traders to continue the old contracts.

That year, the response to the weakening of the tenge did not occur rapidly enough for Kazakhstan to avoid an economic slowdown at that time. Devaluations may take longer to work than policymakers realize. –Leon Taylor, tayloralmaty@gmail.com


Good reading
Paul Krugman and Maurice Obstfeld. International economics: Theory and policy. Eighth edition. 2009.  Chapter 16 discusses the J curve.


References


Stephen Magee. Currency contracts, pass-through, and devaluation. Brookings Papers on Economic Activity. 1973.