Accounting Quality, Uncertainty, and Cost of Debt and Equity Capital in Deregulated Electric Utilities

Electric utilities play an important role in the economy and the general public's life quality, and have caught lots of attention in the literature of Economics. However, the accounting research seems to be lacking in this area. Although the ruling of national deregulation in Electric Industry was announced in 1996, many states opted into the deregulation in early 2000's, completing the first wave of national deregulation. Up to today, only part of the nation's electric market is deregulated and competitive, and the regulated states are still debating whether they should proceed with deregulation. Analysis on the costs and benefits of the deregulation is of great significance to the electric utilities and the public. My paper contributes to this debate by analyzing the benefit of using accounting quality to enhance value creation.In this paper, I use the national deregulation in electric utilities as an exogenous shock for the uncertainty and analyze how accounting quality can help mitigate this uncertainty and affect a firm's cost of debt and equity capital. I find that accounting quality is negatively associated with cost of debt and cost of equity capital after deregulation in electric utilities while such association does not exist between accounting quality and cost of debt and equity in the years before deregulation. Further analysis with various measures of information uncertainty finds that accounting quality mitigates the effects of information uncertainty's on the cost of debt and cost of capital after deregulation, but accounting quality does not mitigates the effects of information uncertainty prior to the deregulation. These results confirm that accounting quality serves as effective instrument to mitigate the effects of information uncertainty on cost of debt and equity capital. I also include various cross-sectional analyses on the deregulated states vs. regulated states, and the results are consistent with my predictions in sign, and larger than the results from above full sample in magnitude