That is the title of a very interesting new paper appearing in Public Administration Review by Qiushi Wang (Sun Yat-sen University) and Jun Peng  (University of Arizona). The paper focuses on discount rate assumptions used in calculating future returns on pension assets. Increasing the discount rate implies that future liabilities are better funded by present assets, or conversely lowering the discount rate makes the pensions appear more unfunded. Actual future market returns are unknown, and funding the pension is in competition with other funding priorities in the current period, so subjective assumptions of the discount rate are at least as much a political consideration as it is a technocratic judgement.

Wang and Peng look at the public pension boards governing state plans and sets up two dimensions for measuring their “political embeddedness” in the Polanyi tradition. The first dimension they look at is the degree to which state government can directly control the outcome of pension decisions through strong ties to the the board (e.g. the use of political appointees). The second dimension is the degree to which pension decisions are influenced by outside interest groups (e.g. public unions) when ties to government are weak. They motivate a series of variables in this framework and estimate a Cox regression to see which factors jump out as being correlated with discount rate revisions between 2002 and 2014.

There is not an enormous amount of variation in the discount rate over time, so generally effect sizes are rather small. Still, the most consistent result they find among the variables motivated by political theory is the percent of political appointees on the board, who seem to be correlated with downward revisions in the discount rate, forcing the plans to appear less funded than before. The presence of a public sector union or having a more conservative state government, by contrast, did not have any effects. Generally speaking, the governments were less likely to make these revisions while experiencing fiscal stress regardless of who was governing them.

Given that public pensions are generally underfunded, these downward revisions are probably likely to be regarded as a more prudent financial change. It is hard to gerrymander an entire state, so it is not particularly clear why political appointees would be more long-term in their orientation. The authors speculate that politically appointed trustees were more concerned about long term health of the pension funds than trustees elected by plan participants. Generally I think people, maybe academics in particular, tend to underrate partisan appointments if there is an explicitly bipartisan effort. So perhaps states with lots of political appointments are making room on the board for bipartisan appointments where they can check each other’s work and enfranchise political allies to garner the trust necessary in defending technocratic assumptions. This is all to say that Wang and Peng have produced some intriguing data and results that are worth reading and pondering.

For some related literature, I would recommend Friedrich Breyer’s piece in the European Journal of Political Economy which attempts to broadly answer the question of why governments attempt to transfer from young to old, and why these transfers are not 100%.