Matthew R. Lyle

Accounting Data, Market Values, and the Cross Section of Expected Returns Worldwide

Under fairly general assumptions, expected stock returns are a linear combination of two accounting fundamentals ― book to market and ROE. Empirical estimates based on this relation predict the cross section of out-of-sample returns in 26 of 29 international equity markets, with a highly significant average slope coefficient of 1.05. In sharp contrast, standard factor-model-based proxies fail to exhibit predictive power internationally. We show analytically and empirically that the importance of ROE in forecasting returns depends on the quality of accounting information. Overall, a tractable accounting-based valuation model provides a unifying framework for obtaining reliable proxies of expected returns worldwide.

Expected Stock Returns Worldwide: A Log-Linear Present-Value Approach

This is the first large-scale study of the performance of expected-return proxies (ERPs) internationally. Analyst-forecast-based ICCs are commonly used to study how policies affect expected returns in international settings, but they are sparsely populated and not robustly associated with future returns. Earnings-model-forecast-based ICCs are well populated, but are unreliable outside the U.S. We provide a solution from a log-linear and present-value (LPV) framework-- combining an accounting valuation anchor, its expected growth, and market prices-- and adapt it to estimate ERPs in a global context. An LPV ERP anchored on the book value of equity is associated with future returns in each of the 29 equity markets we study, and largely subsumes the out-of-sample predictive ability of the most common firm characteristics that have been shown to be associated with expected returns. Our findings also suggest that a firms life-cycle stage provides useful information for ERP estimation.

The Speed of the Market Reaction to Pre-Open versus Post-Close Earnings Announcements

The vast majority of U.S. public firms announce earnings in the post-close (between the closing bell and midnight, or PC) or the pre-open (between midnight and the opening bell, or PO). Prior literature generally treats PC and PO announcements as equivalent when measuring the market reaction to the announcement. In this study, we provide a model of investor processing of earnings announcements and hypothesize that PO announcements have a slower market reaction to the earnings news than PC announcements. This differential speed of the market reaction is because PC announcements offer investors more time to process and trade on the earnings news before regular trading begins. We find strong empirical evidence that, although the earnings news and cumulative market reaction does not differ between PC and PO announcements, PO announcements incorporate earnings news more slowly in the days after the announcement than PC announcements.

Macroeconomic News in the Cross Section of Asset Growth

Firms make forward-looking decisions. We provide evidence that firms’ investment decisions contain news about future aggregate conditions. This information is best extracted by dimension-reduction techniques. The investment-based signal improves upon the widely-used GDP forecasts found in the Survey of Professional Forecasters. We appeal to news-driven business cycle theory to explain our result, suggesting that these investment decisions contain firms’ information about future productivity shocks. This theory also helps us to understand why accounting-based measures of investment reveal the news while market-based measures of value do not.

Firm Fundamentals and Variance Risk Premiums

The same firm characteristics that help explain cross-sectional variation in expected stock returns, such as size, book-to-market and the earnings yield, also help explain cross-sectional variation in returns to trading in option-implied stock return volatility. This empirical phenomenon is shown to arise within a tractable accounting-based valuation model that allows for risk aversion and stochastic earnings volatility. The model predicts that expected stock (stock return volatility) returns are positively (negatively) related to a combination of the inverse of size, book-to-market, the earnings yield, and the dividend yield. These predictions are strongly supported using a variety of empirical specifications. The model provides a framework for jointly investing in stocks and options, and the findings highlight the ability of accounting-based valuation models to explain price dynamics across stock and volatility markets.

Discussion of 'Valuation: Accounting for Risk and the Expected Return'

In this article I discuss Penman (2016), titled “Valuation: Accounting for Risk and the Expected Return.” Penman (2016) is important because it offers potential insights that can help us understand why the book-to-market ratio and other accounting-based variables may impact expected stock returns. It does so by considering the way accounting systems measure assets and income and how these systems deal with risk. My discussion mainly focuses on what Penman calls “Accounting for Risk” and the role of log-linear models in valuation.

How Does Algorithmic Trading Improve Market Quality

We use a comprehensive panel of NYSE order book data to show that the liquidity and quoting efficiency improvements associated with algorithmic trading (AT) are attributable to enhanced monitoring by liquidity providers. We find that variation in liquidity provider monitoring uniquely explains quoting behaviors around idiosyncratic versus multi-asset price jumps and small- versus large-stock price jumps. In addition, we find monitoring outperforms measures of overall AT activity in explaining stock-level decreases in liquidity costs, and that residual variation in AT is associated with increased spreads. Importantly, our results indicate that there are diminishing returns to market function from subsequent technological advancements, thus providing a novel explanation for why spreads have not continued to fall since 2007 despite sustained increases in algorithmic trading.