Body
Beyond Parallel, a project at CSIS, has published what may be the first public opinion poll done inside North Korea. The sample size is small (n=36). And none of the reported results contribute to new insights that have not already been obtained from surveys of North Korean refugees or travelers interviewed outside the country, but one has to hand it to the Beyond Parallel team for sheer audacity.
The main conclusions are that North Koreans don’t believe that they live in socialist paradise and that the public distribution system does not provide an adequate existence. It is generally accepted that the PDS began to falter in the 1980s, and Steph Haggard and I documented how for most of the population it collapsed quickly in the early 1990s. Surveys by the World Food Program and other UN agencies document how today it only delivers about 60% of caloric requirements—and that is on paper. So the fact that North Koreans in 2016 say the PDS doesn’t deliver adequately to fulfill their needs is not surprising.
The fact that the results are so consistent with existing refugee surveys could be interpreted as confirmation that potential sample selection issues with those surveys are not fatal.
Potentially more interesting is the seemingly obvious result that North Koreans do not regard themselves as living in a socialist paradise. They report increasing numbers engaged in informal sector employment, they pay bribes to carry on their businesses, and the botched November 2009 currency reform marked a turning point in turning people away from the regime. Plenty of refugee surveys have documented dissatisfaction, and even precisely these patterns and forms of dissatisfaction. (In fact, Kim Joon-kyung, the president of the Korea Development Institute presented some data consistent with these results yesterday at SAIS.) So what the CSIS work has going for it is that it was conducted in-country so is not subject to the same self-selection biases that the refugee surveys are. The fact that the results are so consistent with existing refugee surveys could be interpreted as confirmation that potential sample selection issues with those surveys are not fatal. That said, the CSIS report is opaque about how the poll was conducted, so it may be subject to its own set of sample biases. It contains direct quotes from the participants, so it seems to have been some kind of semi-structured set of interviews.
So what we have is a small survey, conducted according to an unrevealed methodology, that confirms what many of us thought that we already knew. But the astonishing thing is that it even exists. I look forward to further efforts in this direction.