Does financial disclosure matter? GASB 45 and municipal borrowing costs

ABSTRACT

Financial disclosure is crucial to achieve fiscal transparency and government accountability. In 2004, the Governmental Accounting Standards Board (GASB) issued Statement No. 45 to require the disclosure of information about other postemployment benefits (OPEB). Using a panel of U.S. counties and the bonds they issued between 1999 and 2012, this paper examines the effects of GASB 45 on municipal borrowing costs. GASB 45 increases borrowing costs of county governments, with the effects decreasing over time. GASB 45 has a larger effect on borrowing costs of county governments issuing bonds of lower credit quality and adopting the generally accepted accounting standards (GAAP).

KEYWORDS:

• Financial disclosure
• fiscal transparency
• accountability
• borrowing cost
• GASB
• OPEB

Disclosure statement

No potential conflict of interest was reported by the author.

Supplementary material

Supplemental data for this article can be accessed here.

Notes

1. More information on the setup and estimation of the model is available in the appendix.

2. The estimation omits seven years after GASB 45 implementation because the sample contains too few observations.

3. A table that summarises the names, definitions, and predicted signs of control variables is available in the appendix.

4. As an alternative specification, the model is re-estimated by excluding total expenditures only from the outcome equation. The results remain robust.

5. Because insured bonds carry the ratings of bond insurance firms regardless of their underlying ratings, an alternative specification is to estimate the model for uninsured bonds only. The results are robust.

6. The credit rating by Standard and Poor’s is used because it has the smallest amount of missing values in the sample compared to the ratings by Moody’s or Fitch.

7. The number of bonds in the final sample reduce to 80,416 because of missing values. First, lagging federal fiscal transfer and own source revenue for one year turns all the observations in 1999 into missing. Second, there are missing values for competitive sale and ln of par value in the data from Bloomberg. Third, Bond Buyer Index is missing for bonds issued at some dates. As a robustness check, the Bond Buyer Index in the most recent week is used to replace the missing values in these cases. The main results remain unchanged.

8. The dummy variable indicating year of 2012 is omitted from the estimation because of collinearity.

9. The full results for the control variables in the outcome equation and the selection equation are available upon request.

10. The instrumental variable regression excludes the variable years after GASB 45 because of a lack of instrumental variables. The OLS regression with the same model specification of the 2SLS regression shows an effect of 0.104 (p<0.01) for the variable GASB 45.

Additional information

Notes on contributors

Jinhai Yu

Jinhai Yu is an assistant professor in the School of Public Economics and Administration at Shanghai University of Finance and Economics (SUFE), Shanghai, China. His research examines how policymakers allocate fiscal resources to improve government performance and accountability, focusing on budgeting politics and debt policy. He has published in Public Administration Review, Public Performance and Management Review, and the Journal of Public Policy, among others.