Skip to main content
Log in

The impact of public pension board of trustee composition on state bond ratings

  • Original Paper
  • Published:
Economics of Governance Aims and scope Submit manuscript

Abstract

The declining financial health of public pension systems is increasingly becoming a budgetary concern for many state and local governments. While the academic literature has identified several factors behind the growth in unfunded state and local public pension liabilities, there is mixed evidence on how the composition of a pension system’s board of trustees affects a pension’s financial health. This article contributes to this literature by measuring how public pension board composition affects fund financial health as measured by state bond ratings. With a panel dataset of state pensions between 2001 and 2014 our results indicate that elected board members are consistently associated with lower bond ratings (and thus higher borrowing costs) while appointed and ex-officio board members are associated with higher bond ratings. These results are robust to a number of specifications.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1

Similar content being viewed by others

Notes

  1. Even board of trustee influence on just investment decisions may also affect other areas that impact funding health. For instance, given that most public pensions discount liabilities using their assumed rate of return, investment policies setting the assumed rate of return will also have an impact on funding health.

  2. There is also evidence that credit rating agencies have factored in OPEB obligations into corporate bond ratings (Harper et al. 1991; Maher 1996).

  3. There is also evidence suggesting that credit rating agencies and investors factor in private pension obligations in evaluating corporate bond risk (Carroll and Niehaus 1998; Gopalakrishnan 1994; Martin and Henderson 1983).

  4. Even the Treasury Department’s recent financial plan for Puerto Rico placed pensioners over bondholders (Walsh 2016).

  5. Investor research groups such as Morningstar (Barkley 2012) and financial service providers such as Citi (Hanif et al. 2016) have also issued reports on the pension crisis faced by state and local governments. In response to the variance in reporting across U.S. state and local pensions and the use of liberal discount rates, Moody’s even replaced the Governmental Accounting Standard Board’s (GASB) reporting used by public pensions with their own independent evaluation (Moody’s 2013).

  6. This information is freely available at http://crr.bc.edu/data/public-plans-database/.

  7. This information is freely available at http://www.publicfundsurvey.org/publicfundsurvey/index.htm.

  8. Information is freely available at www.nasra.org.

  9. All locally administered plans were excluded from the analysis. Additionally, data on board composition for Washington State and West Virginia were ambiguous and often contradictory across sources, with only ex-officio numbers shown for West Virginia. Therefore, these states have been largely excluded from the analysis.

  10. There are a total of 263 observations in the sample with a score of “1”.

  11. Importantly, there is significant variation in composition across boards. Specifically, most pensions have a combination of member types and are almost never composed by 100% of a given type. In fact, across the dataset there was only one board made up entirely of elected members, 15 made up entirely of appointed members, and 2 entirely composed of ex-officio members. Additionally, there were 21 pension boards with a supermajority of elected members in 2001 and 34 boards that were either composed of a supermajority of ex-officio or appointed members. In 2014 these numbers remained at 21 and increased to 39 boards respectively.

  12. We also evaluated fixed-effect OLS specifications, with the results largely qualitatively and quantitatively similar to those reported for the above model. For the sake of space these estimates are unreported, though are available upon request.

  13. Along with these reported results, we also considered an alternative specification which weighted each observation by the number of pension plans in each state. Neither the coefficients nor statistical significance were materially different and have thus been excluded from the paper for the sake of space. However, they are available upon request.

  14. Additionally, “% insider” is clearly just (1-“% elected”). While this variable may not provide much additional information in this specification, results are included for completeness.

  15. Calculated as the coefficient multiplied by the number of current notches Moody’s employs.

  16. Calculated as the decline in notches multiplied by 6.87.

  17. Once again, “majority inside” is simply (1-“majority elected”) and while it may not provide much additional information, the coefficients are reported for consistency.

  18. For robustness we also considered subsamples based on the median values of each of the three pension health control variables (funding ratio, ARC, and assumed discount rate). These results, which are excluded for the sake of space though available upon request, indicate no significant material divergence from the results initially obtained. The only exception would be the above-median sample for the ARC corresponding to column 4 of Table  2, which nets the same sign coefficients, but now is significant in every specification (suggesting that appointed and ex-officio members are associated with higher bond ratings). Further, the sub-sample with an above the median assumed discount rate typically results in an opposite sign coefficient and statistical significance relative to that found in column 4 of Table 2, while the results compared to column 4 of Table 3 are statistically insignificant, and comparisons to column 4 of Table 4 suggest sign coefficients that flip. Overall, then our initial findings do tend to be robust to these alternate specifications, though the assumed discount rate may play a role with at least some of these results.

References

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to John A. Dove.

Additional information

The authors would like to thank Connor J. Cosenza for assistance in collecting and compiling much of the data employed in this study.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Dove, J.A., Collins, C.A. & Smith, D.J. The impact of public pension board of trustee composition on state bond ratings. Econ Gov 19, 51–73 (2018). https://doi.org/10.1007/s10101-018-0201-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10101-018-0201-8

Keywords

JEL Classification

Navigation