Dollarization in North Korea

October 4, 2017 6:30 PM

North Koreans have long used foreign currencies as a store of value (indeed, even preferring to hold their savings in gold or trinkets in preference to the domestic currency), and Steph Haggard and I documented the use of foreign currencies for transactional purposes by businesses. A recent paper by Sung Min Mun and Seung Ho Jung, “Dollarization in North Korea: Evidence from a Survey of North Korean Refugees” extends this inquiry, using a 2015 refugee survey to focus on the use of foreign currencies by households.

The term dollarization has its origins in the displacement of local currencies by the US dollar in Latin America during the 1980s. In the North Korean case, both the US dollar and Chinese yuan are used widely, with one of those currencies or the other dominant depending on circumstance. Mun and Jung argue that while the process of dollarization started in the early 1990s, it got a decisive push with the failed currency reform of November 2009. By interviewing refugees who left North Korea in the years immediately before and after this event they are able to examine whether it changed prevailing trends.

The authors examine two widely used indicators of dollarization, the asset substitution index (ASI), a measure of the use of foreign currencies in saving (i.e. as a store of value) and the currency substitution index (CSI), a measure of the use of foreign currencies for transactional purposes (i.e. as a vehicle of exchange).

In the case of North Korea, households generally do not use the formal banking system as a means of holding saving, so Mun and Jung instead calculate ASI on the basis of cash balances. They calculate the measure two ways, as a simple average of CSIs across households, and as an overall share of cash held by the household sector. The former is an indicator of the behavior of the average household; the latter weights households by their wealth.

By 2012, the North Korean won had ceased to be the primary vehicle for exchange by any definition—and this appears to be a permanent change.

Their total sample is 231 individuals (which they treat as 231 households, though they never explicitly state that only one member per household was interviewed). But only about half of the interviewees indicated that they had any saving, so the sample size for ASI calculations is only 118 respondents. But this is large enough that they break the sample not only in pre- and post-currency reform subsamples, but also divide the sample between respondents coming from provinces bordering China and those coming from the rest of the country.

They find a significant jump in the average and overall ASI following the currency reform. In the four years preceding the currency reform, they find that the average ASI was 55.5 (i.e. for the average household, foreign currencies made up a bit more than half of cash balances), but rose to 78.4 in the five years following the event. The overall ASI rose from 71.0 to 97.8, demonstrating that wealthier households were more dollarized pre-reform, and after the failed currency reform, the North Korean won effectively ceased being used as a vehicle for household saving.

So what replaced the North Korean won in household saving? Before the reform in the border region 48.0 percent of saving was held in yuan and 9.3 percent in dollars. However, in the rest of the country, 32.0 percent of saving was held in dollars, and 27.8 percent was held in yuan. After the currency reform, in the border region, most household saving was held in yuan (78.3 percent) while in the interior, most was held in dollars (50.4 percent).

Similar patterns are evident with respect to transactional use. In the four years prior to the reform, CSI averaged 25.3; in the five years after the reform it rose to 51.4—meaning that the North Korean won had ceased to be the primary currency for transaction for the average household. In terms of overall CSI, pre-reform it was 40.8 (again reflecting higher rates of foreign currency use among wealthier households), afterwards rising to 67.0. From 2012 through the end of the sample in 2015, both CSI measures are above 50, indicating that by 2012, the North Korean won had ceased to be the primary vehicle for exchange by any definition—and this appears to be a permanent change. Again, the Chinese yuan predominates in the border region, and the US dollar elsewhere.

There is variation in currency use by type of transaction. The North Korean won still is used most often in the purchases of rice, corn, and flour. But market sales of grain remain technically illegal, and state-run distribution channels are more prominent for these items. For non-food items such as electronics, housing, and loans, foreign currencies predominate post-reform.

Mun and Jung then ask the question if North Korea is running a trade deficit, how are these foreign currencies being acquired? Their answer, citing research by Haggard and Noland, is that North Korea has non-trade sources of revenue, and at times may have even run a current account surplus. But there is another explanation which they miss. To calculate these ratios, one has to convert the figures into a common currency. Mun and Jung correctly use the (black-) market exchange rate, not the wildly overvalued official rate to do this conversion. The issue is that for much of their sample period, the North Korean won depreciated, particularly after the failed currency reform. So from a purely accounting standpoint, the prominence of foreign transactions would appear to grow, even if households were unable to acquire physically more dollars or yuan. This valuation effect could also explain why dollarization appears to continue apace despite the attempts by the state to extract foreign currency from the household sector that the authors describe.

The policy issue that remains unexamined is how the effective dollarization of the economy constrains the regime’s macroeconomic policy by effectively neutering monetary policy. That remains for another post. Mun and Jung have usefully fleshed out our knowledge of foreign currency use by the household sector.

My original intent was to use "If Money Talks" as the closing video. But under the circumstances...RIP Tom Petty.


Add new comment