Working Papers by Richard Roll
# | Title | Authors | Date | Length | Paper | Abstract | |
---|---|---|---|---|---|---|---|
1463 | Competition Shocks, rival reactions and return comovement | Roll, Richard de Bodt, Eric Eckbo, B. Espen | 03/21/2022 | 51 | sswp1463.pdf | We estimate changes in within-industry stock-return comovement caused by the reaction of rival firms to significant tariff cuts. In theory, rivals react by either increasing or decreasing product differentiation. Increased differentiation lowers cash flow correlation and return comovement, while reduced differentiation increases comovement. Large-sample tests show that tariff cuts in manufacturing industries increase comovement and more so for within-industry ‘followers' than ‘leaders'. The notion that this comovement-increase reflects efficiency-enhancing rival reactions is also supported by evidence of increased cost-efficiency measures. One channel for this efficiency-increase is M&A activity among industry followers. | |
1462 | The (Un)intended Consequences of M&A Regulatory Enforcements | Roll, Richard de Bodt, Eric Cousin, Jean-Gabriel Officer, Micah | 03/21/2022 | 56 | sswp1462.pdf | Economic and policy uncertainty affect merger and acquisition (M&A) activity. In this paper, we use Department of Justice (DOJ) and Federal Trade Commission (FTC) interventions in the M&A market to investigate whether uncertainty around regulatory enforcements also matters. Our results support this conjecture. Using the Hoberg and Phillips (2010) similarity scores to identify product market competitors, we confirm a clear and significant DOJ/FTC regulatory enforcements' deterrence effect on future M&A transaction attempts, a result robust to many alternative specifications and confirmed in additional tests. This deterrence effect is (at least partly) driven by the length of the regulatory process, a factor that exacerbates enforcement uncertainty. Our results identify an (un)intended channel through which M&A regulation hampers efficient resources allocation. | |
1456 | The Efficient Frontier: A Note on the Curious Difference Between Variance and Standard Deviation | Roll, Richard | 07/01/2021 | 9 pages | sswp_1456.pdf | The Markowitz Frontier of optimal portfolios is valid in both mean/variance space and in mean/standard deviation space. But there are some curious differences because lines in one space become curves in the other. This note explores and explains the curiosity. | |
1446 | Changing Expected Returns Can Induce Spurious Serial Correlation | Pukthuanthong, Kuntara Roll, Richard Subrahmanyam, Avanidhar | 09/21/2021 | 59 | sswp1446_revised.pdf | Changing expected returns can induce spurious autocorrelation in returns. We show why this happens with simple examples and investigate its prevalence in actual equity data. In a key contribution, we use ex ante expected return estimates from options prices, factor models, and analysts' price targets to investigate our premise. Absolute shifts in expected returns are indeed strongly and positively related to autocorrelations in the cross-section of individual stocks, as predicted by our analysis. Well-studied risk factors show no evidence of spurious components. We also show how our analysis implies spurious cross-autocorrelation and find supporting evidence for this phenomenon as well. | |
1436 | Mimicking Portfolios | Roll, Richard Srivastava, Akshay | 01/08/2018 | 29 | sswp1436.pdf | Mimicking portfolios have many applications in the practice of finance. Here, we present a new method for constructing them. We illustrate its application by creating portfolios that mimic individual NYSE stocks. On the construction date, a mimicking portfolio exactly matches its target stock's exposures (betas) to a set of ETFs, which serve as proxies for global factors, and the portfolio has much lower idiosyncratic volatility than its target. Mimicking portfolios require only modest subsequent rebalancing in response to instabilities in target assets and assets used for portfolio construction. Although composed here exclusively of equities, mimicking portfolios show potential for mimicking non-equity assets as well. | |
1435 | Tick Size, Price Grids and Market Performance: Stable Matches as a Model of Market Dynamics and Equilibrium | Plott, Charles R. Roll, Richard Seo, Han Zhao, Hao | 01/08/2018 | 58 | sswp1435_r_X1x1ZD3.pdf | This paper reports experiments motivated by ongoing controversies regarding tick size in markets. The minimum tick size in a market dictates discrete values at which bids and asks can be tendered by market participants. All transaction prices must occur at these discrete values, which are established by the rules of each exchange. The simplicity of experiments helps to distinguish among competing models of complex real-world securities markets. We observe patterns predicted by a matching (cooperative game) model. Because a price grid damages the equilibrium of the competitive model, the matching model provides predictions where the competitive model cannot; their predictions are the same when a competitive equilibrium exists. The experiment examines stable allocations, average prices, timing of order flow and price dynamics. Larger tick size invites more speculation, which in turn increases liquidity. However, increased speculation leads to inefficient trades that otherwise would not have occurred. | |
1431 | A Protocol for Factor Identification | Pukthuanthong, Kuntara Roll, Richard Subrahmanyam, Avanidhar | 07/28/2017 | 51 | sswp1431.pdf | We propose a protocol for identifying genuine risk factors. The underlying premise is that a risk factor must be related to the covariance matrix of returns, must be priced in the cross-section of returns, and should yield a reward-to-risk ratio that is reasonable enough to be consistent with risk pricing. A market factor, a profitability factor, and traded versions of macroeconomic factors pass our protocol, but many characteristic-based factors do not. Several of the underlying characteristics, however, do command material premiums in the cross-section. | |
1430 | Generalized Portfolio Performance Measures: Optimal Overweighting of Fees Relative to Sample Returns | Levy, Moshe Roll, Richard | 07/27/2017 | 24 | sswp1430.pdf | Performance measures such as alpha and the Sharpe ratio are typically based on sample returns net of fees. This implies the same weighting to sample returns and to fees. However, sample return parameters are noisy estimates of true parameters, while fees are known with certainty. Thus, intuition suggests that fees should be given more weight than sample returns. We formalize this intuition, and derive the optimal overweighting of fees. We show that the resulting generalized performance measures are better predictors of future net performance than the standard performance measures, and they better explain future fund flows. | |
1426 | Empirical Evidence of Overbidding in M&A Contests | de Bodt, Eric Cousin, Jean-Gabriel Roll, Richard | 11/15/2016 | 61 | Empirical_Evidence_of_Overbidding_in_MA.pdf | Surprisingly few papers have attempted to develop a direct empirical test for overbidding in M&A contests. We develop such a test grounded on a necessary condition for profit maximizing bidding behavior. The test is not subject to endogeneity concerns. Our results strongly support the existence of overbidding. We provide evidence that overbidding is related to conflicts of interest, but also some indirect evidence that it arises from failing to fully account for the winner's curse. | |
1425 | Nowhere to Run, Nowhere to Hide: Asset Diversification in a Flat World | Cotter, John Gabriel, Stuart Roll, Richard | 10/20/2016 | 58 | SSWP_1425.pdf | We present new international diversification indexes across equity, sovereign debt, and real estate. The indexes reveal a marked and near ubiquitous decline in diversification potential across asset classes and markets for the post-2000 period. Analysis of panel data suggests that the decline is related to higher levels of market credit risk and volatility as well as to technology and communications innovation as proxied by internet diffusion. The decline in diversification opportunity is associated with sharply higher levels of investment risk. | |
1419 | Improved Methods for Detecting Acquirer Skills | de Bodt, Eric Cousin, Jean-Gabriel Roll, Richard | 05/17/2016 | 42 | SSWP1419.pdf | Large merger and acquisition (M&A) samples feature the pervasive presence of repetitive acquirers. They offer an attractive empirical context for revealing the presence of acquirer skills (persistent superior performance). But panel data M&A are quite heterogeneous; just a few acquirers undertake many M&As. Does this feature affect statistical inference? To investigate the issue, our study relies statistical support for the presence of acquirer skills appears compromised. We introduce a new resampling method to detect acquirer skills with attractive statistical properties (size and power) for presence of acquirer skills but only for a marginal fraction of the acquirer population. This result is robust to endogenous attrition and varying time periods between successive transactions. Claims according to which acquirer skills are a first order factor explaining acquirer cross-sectional cumulated abnormal returns appears overstated. | |
1417 | Full Stock Payment Marginalization in M&A Transactions | de Bodt, Eric Cousin, Jean-Gabriel Roll, Richard | 04/05/2016 | 55 | SSWP_1417.pdf | The number of merger and acquisition (M&A) transactions paid fully in stock in the U.S. market declined sharply after 2001, when pooling and goodwill amortization were abolished by the Financial Accounting Standards Board. Did this accounting rule change really have such far reaching implications? Using a differences-in-differences test and Canada as a counterfactual, this study reveals that it did. We also report several other results confirming the role of pooling abolishment, including (i) that the decrease in full stock payment relates to CEO incentives and (ii) that previously documented determinants of the M&A mode of payment cannot explain the post pooling abolishment pattern. These results are also robust to controls for various factors, such as the Internet bubble, the exclusion of cross-border deals, the presence of Canadian cross-listed firms, the use of a constant sample of acquirers across the pooling and post pooling abolishment periods, the use of Europe as an alternative counterfactual, and controls for the SEC Rule 10b-18 share repurchase safe harbor amendments of 2003. | |
1411 | Seeking Alpha? It's a Bad Guideline for Portfolio Optimization | Levy, Moshe Roll, Richard | 12/10/2015 | 18 | SSWP_1411.pdf | Alpha is the most popular measure for evaluating the performance of both individual assets and funds. The alpha of an asset with respect to a given | |
1405 | An Agnostic and Practically Useful Estimator of the Stochastic Discount Factor | Pukthuanthong, Kuntara Roll, Richard | 03/05/2015 | 72 | SSWP_1405R.pdf | We propose an estimator for the stochastic discount factor (SDF) which is agnostice because it does not require macroeconomic proxies or preference assumptions. It depends only on observed asset returns. Nonetheless, it is immune to the form of the multivariate return distribution, including the distribution's factor structure. Putting our estimator to work, we find that a unique positive SDF prices all U.S. asset classes and satisfies the Hansen/Jagannathan variance bound. In contrast, the Chinese and Indian equity markets do not share the same SDF and hence do nto seem to be integrated. | |
1398 | Can Housing Risk be Diversified? A Cautionary Tale from the Housing Boom and Bust | Cotter, John Gabriel, Stuart Roll, Richard | 10/13/2014 | 54 | SSWP_1398.pdf | The study evaluates the effectiveness of geographic diversification in reducing housing investment risk. To characterize diversification potential, we estimate spatial correlation and integration among 401 US metropolitan housing markets. The 2000s boom brought a marked uptrend in housing market integration associated with eased residential lending standards and rapid growth in private mortgage securitization. As boom turned to bust, macro factors including employment and income fundamentals, contributed importantly to the trending up in housing return integration. Portfolio simulations reveal substantially lower diversification potential and high risk in the wake of increased market integration. | |
1393 | The Propagation of Shocks Across International Equity Markets: A Microstructure Perspective | Roll, Richard Bongaerts, Dion Rosch, Dominik van Dijk, Mathijs Yuferova, Darya | 08/12/2014 | 53 | SSWP1393.pdf | We study the high-frequency proagation of shocks across international equity markets. We identify shocks to stock prices, liquidity (quoted and effective spreads), and trading activity (turnover and order imbalance) for 12 equity markets around the world based on non-parametric jump statistics at the 5-minute frequency from 1996 to 2011. Jumps in prices, quoted spreads, and order imbalance are prevalent and large, while jumps in effective spreads and turnover are rare. Within a market, jumps in prices are regularly accompanied by jumps in order imbalance, but are independent of jumps in liquity. Jumps in prices and co-jumps in prices and order imbalance are primarily driven by information rather than liquidity, since there is no subsequent price reversal and since they often occur around U.S. macroeconomic news announcements. We also present evidence that jumps in prices and order imbalance (but not liquity) spillover across markets within the same 5-minute interval. Overall, we find that order imbalances help explain why shocks to stock prices occur and spread across markets, while shocks to liquity tend to be isolated events. | |
1392 | Resolving the Errors-in-Variables Bias in Risk Premium Estimation | Roll, Richard Wang, Junbo Pukthuanthong, Kuntara | 07/21/2014 | 57 | SSWP_1392.pdf | The Fama-Macbeth (1973) rolling-B method is widely used for estimating risk premiums, but its inherent errors-in-variables bias remains an unresolved problem, particularly when using individual assets or macroeconomic factors. We propose a solution with a particular instrumental variable, B calculated from alternate observations. The resulting estimators are unbiased. In simulations, we compare this new approach with several existing methods. The new approach corrects the bias even when the sample period is limited. Moreover, our proposed standard errors are unbiased, and lead to correct rejection size in finite samples.
| |
1390 | The Hubris Hypothesis: Emperical Evidence | Roll, Richard de Bodt, Eric Cousin, Jean-Gabriel | 07/10/2014 | 49 pages | SSWP_1390.pdf | The Hubris Hypothesis is grounded on a failure to adequately account for the winner's curse, which leads to overbidding. Surprisingly, few papers have attempted to develop a direct empirical test of the presence of overbidding of M&A contests. We develop two such tests in this paper. Our results strongly support the existence of overbidding.
|