Association for Public Policy Analysis & Management

This paper aims to explore how different types of volatility risk measures influence the diversification of nonprofit organizations’ revenue portfolios. The literature on financial risk management in the nonprofit sector has been dominated by a focus on deviation risk estimations using the coefficient of variance (CV) and on its relationship with diversification using the Herfindahl-Hirschman Index (HHI). This paper, however, has adopted alternative methods of those measures, including value at risk (VaR) instead of CV for the revenue volatility measure and the portfolio variance index instead of HHI for the portfolio diversification measure.

The CV is a useful tool for measuring financial volatility, in general. It tells us how the expected value is relatively dispersed from the mean. For example, a low standard deviation indicates that the data point tends to be close to the mean, whereas a high standard deviation indicates that the data point tends to be placed far from the mean. The major issue of this measure presents both upside and downside risks. If nonprofit managers are risk-averse, nonprofit managers may be concerned more with the downside risk than the aggregate deviation risk, which includes both upside and downside risks. Therefore, this paper discusses an alternative method of measuring financial risk in the nonprofit sector, focusing on downside volatility risk (the probability of loss) by implementing VaR with the non-parametric kernel density estimate.

VaR measures the downside risk and quantifies outcomes that occur in the lower tail of statistical probability. More specifically, VaR answers the question of how much one could lose, typically with a 95- or 99-percentile probability. The non-parametric kernel density estimate helps to calculate VaR when the data do not follow the normal probability distribution or have an unknown distribution. The CV as a volatility measure might be inaccurate in the case of a non-normal distribution, as the CV is rooted in the central limit theorem and it requires a normal probability distribution.

The portfolio variance-covariance matrix has been used to measure the divarication of nonprofit organization’s portfolios to compensate for the weakness of the HHI. The HHI is one of the most common measures of portfolio concentration, but it might ignore the interaction across assets and be difficult to interpret as financial risk concentrations.

This paper finds that downside risk does not influence portfolio diversification, whereas deviation risk has an inverse relationship with portfolio diversification. We establish models that account for the factors affecting volatility risk and the diversification of portfolio structure, such as organizational characteristics, governance, and financial characteristics. The models tested multiple observations of service fields based on the National Taxonomy of Exempt Entities (NTEE) over a period of five years from 2008 to 2012 by employing the fixed-effects regression model.

This paper will help nonprofit managers and scholars to distinguish downside risk from deviation risk estimation and to understand the importance of accurate risk estimation and portfolio management. Moreover, this paper will fit into a conference track, Public and Nonprofit Management and Finance.