(April 8, 2013) - Looking to make an allocation to infrastructure? Find a small, highly leveraged firm with low cash flow volatility, according to new study from the researchers in Switzerland.
In "Common Risk Factors for Infrastructure Firms," University of St. Gallen finance scholars Martin Eling and Semir Ben Ammar posed two new models for investors to better evaluate the asset class. In total, the authors tested the impact of 10 factors on 32 years of listed infrastructure returns: leverage, cash flow volatility, capital investment growth, term risk, default risk, regulatory risk, momentum, firm size, price-to-book ratio, and market risk.
The most notable result was with leverage: the authors found a strong, economically significant positive relationship between the extent to which an infrastructure firm was leveraged and the performance of its shares. This study, according to the authors, marked the first empirical investigation of the relationship between leverage and infrastructure returns.
Market performance, default risk, term risk, and, to a minor degree, momentum, also showed statistically significant positive relationships to infrastructure returns.
As for negative relationships of consequence, the size of firms, cash flow volatility, and invested capital flow all tended to track inverse to the performance of the asset class.
Price-to-book ratio and regulation risk showed no significant link.
The results could be useful to both investors and policy makers, the authors contended, as both will likely have substantial allocation power and exposure in the industry going forward. "[As] factor loadings present the correlation structure between returns and the underlying risk factors, factor exposures indicate strategies for minimum variance hedging, for the replication of infrastructure returns, or, in general, for optimizing asset allocation," the authors wrote.
Eling and Ben Ammar's data set used in the study was extensive, but attested to the early stage of the asset class's development. Their data spanned from 1980 to the close of 2011, making it one of the most comprehensive studies of infrastructure investing to have been done. However, it is comprised of financial data on listed corporations that own or operate physical infrastructure assets, not the performance of direct assets themselves.
Listed infrastructure provided both reliable and long-term data, the authors explained. "Further, unlike equity infrastructure indices, we neither determine a specific liquidity level nor a certain market capitalization for our dataset. This means we are able to analyze the entire scope of infrastructure stocks without limitations."
Eling and Ben Ammar did, however, suggest the value of a similar study using unlisted projects.
Read the entire paper here.