Common Risk Factors of Infrastructure Firms (joint work with M. Eling)

Transcription

1 (joint work with M. Eling) Semir Ben Ammar University of St. Gallen, Switzerl 2013 ARIA Annual Meeting Washington, DC August 5 th 2013

2 2 Agenda

3 3 Infrastructure Investments become more more relevant. Low interest rates Insurance companies searching for alternatives Great need for infrastructure investments (OECD, 2007): $70,000bn until 2030 Infrastructure: A separate asset class?(!) Contribution Derive a factor model to explain the variation in infrastructure returns Cost of capital determination hedging possibilities One of the most comprehensive datasets in infrastructure research (1/1980 to 12/2011; subsectors; b/m size differentiation) Identification of underlying risk factors of infrastructure investments Relativization of assumed benefits of infrastructure investments Superior performance (in terms of alpha adj. R-square) compared to established equity factor models First analysis of infrastructure regarding interest rate sensititvity (ALM)

4 4 Assumed benefits what are the major findings until now? Low risk in terms of market beta but overall similar volatility as market index (Rothballer Kaserer, 2012) Inflation hedge hedge seems to be rather «wishful thinking» than reality (Rödel Rothballer, 2012) Stable cash flows infrastructure funds show more stable cashflows than non-infrastructure funds cash flow volatility is valued at a premium (Bitsch, 2012) Uncorrelated returns low correlations with traditional asset classes in Australia (Finkenzeller et al., 2010) Downside protection no downside protection (Bird et al., 2012) vs. potential downside protection through direct infrastructure (Dechant Finkenzeller, 2012)

5 5 Hypothesis Rationale of Infrastructure Variables Variable Hypothesis Rationale Cash flow volatility (CFVOLA) Leverage (LEV) Investment growth (INV) Term premium (TERM) Default premium (DEF) Regulatory premium (REG) Negative relation between cash flow volatility infrastructure returns. (-) Positive relation between leverage infrastructure returns. (+) Positive relation between investment growth infrastructure returns. (+) Positive relation between term structure infrastructure returns. (+) Positive relation between default premium infrastructure returns. (+) Positive/negative relation between regulatory risk infrastructure returns. (+/-) Long-term concessions inelastic dem should be valued at a premium. High leverage implies higher risk, especially due to the fact that infrastructure firms are already highly leveraged. Infrastructure firms that invest in large-scale projects renew their physical assets are able to generate excess returns in the long run. Changes in long-term interest rates might reduce value of debt. Credit risk needs to be compensated On the one h: rates of return. Investors expect a compensation rates of return regulation. On the other h: Government can act as a lender of last resort help infrastructure firms in distressed times.

6 6 Construction of Factors All factors are excess returns! Cashflow volatility factor Rolling stard deviation over past 36 months of quarterly data. Return difference between high low cashflow volatility stocks. Yearly rebalanced. Leverage factor Return difference between high leverage low leverage stocks. Investment factor Construction as in Chen et al. (2011). Low investment minus high investment stocks. Term structure Long-term government bond index minus 1-month T-Bill. Default premium Long-term corporate bond index (BAA) minus long-term government bond index. Regulatory premium Utility bond index minus Industrial bond index. Market factor, SMB, HML MOM from Kenneth French data library.

7 7 Infrastructure Model as an augmented a reduced Version 10-factor augmented version on Carhart (1997) 4-factor model: 7-factor reduced version with infrastructure-specific factors only:

8 8 Data Portfolios follow previous findings about stock returns Monopolistic pattern unlikely Lower stard deviation than market index but also slightly lower return

9 9 Main Results Higher explanatory power of 7-10-factor models. All infrastructure-specific factors except for investment factor statistically significant. Leverage absorbs relevance of HML.

10 10 Alpha adjusted R-square Alpha Adj. R-square CAPM Fama/French Infrastructure 7-factor Infrastructure 10-factor

11 11 Subsectors are rather heterogeneous Heterogeneous exposure Leverage more relevant than HML Market factor most important factor for non-utilities

12 Fittet infrastructure returns from 10-factor model 12 Defensive Characteristics of Infrastructure Firms Only one portfolio shows signs of downside protection Downside protection cannot be assumed for infrastructure firms (corroborates of Bird et al., 2012) Market excess return

13 13 Inflation Hedge Orthogonalization of inflation factor Weighted average infrastructure index does not hedge against inflation Low B/M portfolios utility firms exhibit weak inflation hedge

14 14 Subperiods Out-of-Sample Term Cashflow volatility are the most consistent factors over time Out-of-sample values are overall moving in the same range

15 15 Commercial Infrastructure Indices

16 16 Leverage has a greater impact on the returns of infrastructure firms than HML Cashflow stability is valued at a premium Infrastructure firms are indeed interest rate sensitive (in terms of term structure credit spread) The infrastructure sector is heterogeneous, loading differently on the factors. Asset managers can hedge infrastructure firms with different loadings on the risk factor to minimize their risk exposure Infrastructure hedges inflation only weakly in the utilities sector with low B/M ratio Downside protection is unlikely

17 17 Literature Bird, R., Liem H. Thorp, S., Infrastructure: Real assets real returns, European Financial Management, 2012, doi: /j X x. Bitsch, F., Do investors value cash flow stability of listed infrastructure funds? Working Paper (TU Munich, 2012). Carhart, M. M., On persistence in mutual fund performance, Journal of Finance, Vol. 52, 1997, pp Chen, L., Novy-Marx, R. Zhang, L., An alternative three-factor model, Working Paper (Washington University in St. Louis, 2011). Dechant, T. Finkenzeller, K., The role of infrastructure investments in a multi-asset portfolio Answers from dynamic asset allocation, Working Paper (University of Regensburg, 2012). Fama, E. F. French, K. R., Common risk factors in the returns on stocks bonds, Journal of Financial Economics, Vol. 33, 1993, pp Finkenzeller, K., Dechant, T. Schäfers, W., Infrastructure: A new dimension of real estate? An asset allocation analysis, Journal of Property Investment Finance, Vol. 28, 2010, pp Rödel, M. Rothballer, C., Infrastructure as hedge against inflation fact or fantasy?, Journal of Alternative Investments, Vol. 15, 2012, pp Rothballer, C. Kaserer, C., The risk profile of infrastructure investments: Challenging conventional wisdom, Journal of Structured Finance, Vol. 18, 2012, pp