Equity Premium

A subscriber suggested review of the February 2017 paper “Bubbles for Fama”, in which Robin Greenwood, Andrei Shleifer and Yang You assess Eugene Fama’s claim that stock prices do not exhibit bubbles. They define a bubble candidate as a value-weighted U.S. industry or international sector that rises over 100% in both raw and net of market returns over the prior two years, as well as 50% or more raw return over the prior five years. They define a crash as a 40% drawdown within a two-year interval. They also look at characteristics of industry/sector portfolios identified bubble candidates, including level and change in volatility, level and change in turnover, firm age, return on new versus old companies, stock issuance, book-to-market ratio, sales growth, price-earnings ratio and price acceleration (abruptness of price run-up). They evaluate timing strategies that switch from an industry portfolio to either the market portfolio or cash (with risk-free yield) based on a price run-up signal, or a signal that combines price run-up and other characteristics. Their benchmark is buying and holding the industry portfolio. Using value-weighted returns for 48 U.S. industries (based on SIC code) during January 1926 through March 2014 and for 11 international sectors (based on GICS codes) during October 1985 through December 2014, they find that:

Keep Reading

What happens if we extend the “Simple Asset Class ETF Value Strategy” (SACEVS) with a utilities risk premium, derived from the yield on Utilities Select Sector SPDR Fund (XLU)? To investigate, we apply the SACEVS methodology to the following asset class exchange-traded funds (ETF), plus cash:

3-month Treasury bills (Cash)
iShares 20+ Year Treasury Bond ETF (TLT)
iShares iBoxx $ Investment Grade Corporate Bond ETF (LQD)
XLU
SPDR S&P 500 ETF Trust (SPY)

This set of ETFs relates to four risk premiums, as specified below: (1) term; (2) credit (default); (3) utilities; and, (4) equity. We focus on effects of adding the utilities risk premium on gross compound annual growth rates (CAGR), maximum drawdowns (MaxDD) and annual Sharpe ratios of the Best Value (picking the most undervalued premium) and Weighted (weighting all undervalued premiums according to degree of undervaluation) versions of SACEVS. Using lagged quarterly S&P 500 earnings, monthly S&P 500 Index levels and monthly yields for 3-month U.S. Treasury bill (T-bill), the 10-year Constant Maturity U.S. Treasury note (T-note), Moody’s Seasoned Baa Corporate Bonds since March 1989 (limited by availability of earnings data), XLU prices and dividends since December 1998 (inception) and monthly dividend-adjusted closing prices for the above asset class ETFs since July 2002, all through May 2024, we find that: Keep Reading

Does the S&P 500 Index (SPX) or the CBOE Volatility Index (VIX) yield better short-term trading signals for stocks and VIX futures? In the May 2024 revision of his paper entitled “Chicken and Egg: Should you use the VIX to time the SPX? Or use the SPX to time the VIX?”, Robert Hanna explores mutual predictive relationships between SPX and VIX, with an eye toward exploitation via market timing strategies. He considers several long-term trend indicators to investigate whether SPX or VIX data offers better SPX return predictions. He considers two types of short-term overbought/oversold predictive rules: (1) short-term relative strength index (RSI) readings of 2, 3 and 4 days; and, (2) short-term high and low readings of 5 to 25 days in length. He applies both sets of short-term rules separately to SPX and VIX to predict movements of SPX and VIX futures. Using daily SPX and VIX levels since 1990 and short-term VIX futures prices since 2007, all through 2023, he finds that: Keep Reading

“Countercyclical Asset Allocation Strategy” summarizes research on a simple countercyclical asset allocation strategy that systematically raises (lowers) the allocation to an asset class when its current aggregate allocation is relatively low (high). The underlying research is not specific on calculating portfolio allocations and returns. To corroborate findings, we use annual mutual fund and exchange-traded fund (ETF) allocations to stocks and bonds worldwide from the 2024 Investment Company Fact Book data tables to determine annual countercyclical allocations for stocks and bonds (ignoring allocations to money market funds). Specifically:

• If actual aggregate mutual fund/ETF allocation to stocks in a given year is above (below) 60%, we set next-year portfolio allocation below (above) 60% by the same percentage.
• If actual aggregate mutual fund/ETF allocation to bonds in a given year is above (below) 40%, we set next-year portfolio allocation below (above) 40% by the same percentage.

We then apply next-year allocations to stock (Fidelity Fund, FFIDX) and bond (Fidelity Investment Grade Bond Fund, FBNDX) mutual funds that have long histories. Based on Fact Book annual publication dates, we rebalance at the end of April each year. Using the specified actual fund allocations for 1984 through 2023 and FFIDX and FBNDX May through April total returns and end-of-April 1-year U.S. Treasury note (T-note) yields for 1985 through 2024, we find that: Keep Reading

Are the industry membership of a firm, as designated by Standard Industrial Classification (SIC) code, and the position of the firm within its industry good predictors of the performance of its stock? In their May 2024 paper entitled “Decoding Cross-Stock Predictability: Peer Strength versus Firm-Peer Disparities”, Doron Avramov, Shuyi Ge, Shaoran Li and Oliver Linton devise the following two industry related stock metrics and test their abilities to predict stock returns:

1. Peer Index (PI) – calculated for each firm via a multi-input, inception-to-date regression to predict next-month stock return, replacing firm characteristics by the contemporaneous average values for all firms in its industry as inputs.
2. Peer-Deviation Index (PDI) – calculated for each firm via a multi-input, inception-to-date regression to predict next-month stock return using firm characteristics minus the contemporaneous average values of these characteristics for all firms in its industry as inputs (indicating the standing of the firm within its industry).

Inputs consist of 94 firm-specific characteristics and 8 industry-related characteristics, organized into six groups: momentum, value versus growth, investment, profitability, trading frictions and intangibles. Using monthly values for the selected 102 firm/industry characteristics and monthly returns for common stocks in the top 80% of AMEX/NYSE/NASDAQ  market capitalizations during January 1980 through March 2022, they find that: Keep Reading

What happens if we extend the “Simple Asset Class ETF Value Strategy” (SACEVS) with a real estate risk premium, derived from the yield on equity Real Estate Investment Trusts (REIT), represented by the FTSE NAREIT Equity REITs Index? To investigate, we apply the SACEVS methodology to the following asset class exchange-traded funds (ETF), plus cash:

3-month Treasury bills (Cash)
iShares 20+ Year Treasury Bond (TLT)
iShares iBoxx $ Investment Grade Corporate Bond (LQD)
SPDR Dow Jones REIT (RWR) through September 2004 dovetailed with Vanguard REIT ETF (VNQ) thereafter
SPDR S&P 500 (SPY)

This set of ETFs relates to four risk premiums, as specified below: (1) term; (2) credit (default); (3) real estate; and, (4) equity. We focus on effects of adding the real estate risk premium on gross compound annual growth rates (CAGR), maximum drawdowns (MaxDD) and annual Sharpe ratios of the Best Value (picking the most undervalued premium) and Weighted (weighting all undervalued premiums according to degree of undervaluation) versions of SACEVS. Using lagged quarterly S&P 500 earnings, monthly S&P 500 Index levels and monthly yields for 3-month U.S. Treasury bill (T-bill), the 10-year Constant Maturity U.S. Treasury note (T-note), Moody’s Seasoned Baa Corporate Bonds and FTSE NAREIT Equity REITs Index since March 1989 (limited by availability of earnings data), and monthly dividend-adjusted closing prices for the above asset class ETFs since July 2002, all through May 2024, we find that: Keep Reading

How should investors think about the correlation between stock market returns and bond market returns when constructing a diversified portfolio? In their April 2024 paper entitled “Stock-Bond Correlation: Theory & Empirical Results”, Lorenzo Portelli and Thierry Roncalli examine theoretical and empirical relationships between the stocks-bonds return correlation and other variables/conditions. They focus on:

• Monthly returns of long-term government bonds and country stock markets (for example 10-year U.S. Treasury notes and the S&P 500 Index for the U.S.), but consider other choices for bond duration and stock portfolio.
• 48-month rolling returns when assessing correlation dynamics.

Using government bond and country stock market returns for the U.S. during 1965 through 2023 and for other developed, developing and emerging markets across Europe, the Americas and Asian as available through 2023, they find that:

Keep Reading

Are sentiments conveyed in Seeking Alpha articles useful for stock picking? In their January 2023 paper entitled “Seeking Alpha: More Sophisticated Than Meets the Eye”, Duo Selina Pei, Abhinav Anand and Xing Huan apply two-pass natural language processing to test the informativeness of articles from Seeking Alpha incremental to publicly available earnings data. Specifically, they each month:

• Associate articles with one or more specific stocks.
• Extract positive and negative sentiment at both phrase and aggregate levels for each article/stock.
• Calculate a standardized net sentiment for each article/stock based on the difference between positive and negative mentions, emphasizing event sentiment over general sentiment.
• Rank articles/stocks based on standardized net sentiment over the last month. Reform equal-weighted portfolios of articles/stocks by ranked tenths (deciles). Calculate both immediate [-1,+1] and 90-day future [+2,+90] average gross raw returns and average gross abnormal returns adjusted for size, book-to-market and momentum.
• Sort stocks into 20 groups based on monthly standardized net sentiments up to two days before portfolio selection, excluding stocks with few articles or neutral sentiment. Reform an equal-weighted hedge portfolio that is long stocks with the highest sentiments and short stocks with the lowest (on average, 105 long and 86 short positions).

Using 350,095 articles published on Seeking Alpha since its inception in 2004 through the beginning of October 2018, daily returns of matched stocks and their options and associated earnings surprise data as available, they find that: Keep Reading

Does pricing of options on leveraged exchange-traded funds (ETF) predict future returns of the underlying 1X ETFs? In the March 2024 version of their paper entitled “Lever Up! An Analysis of Options Trading in Leveraged ETFs”, Collin Gilstrap, Alex Petkevich, Pavel Teterin and Kainan Wang examine options trading in leveraged equity ETFs and its implications for future performance of underlying funds. They hypothesize that the compounded leverage of such options attracts especially sophisticated investors. Specifically, they test a risk-on/risk-off strategy that, at the end of each month:

1. Calculates the difference in changes in implied volatilities between at-the-money (ATM) call options and ATM put options on a leveraged ETF (and separately for comparison, on its underlying 1X ETF).
2. If this difference is greater (smaller) than its median value over the prior 12 months, specifies the next month as bullish (bearish) for the 1X ETF, and invests in a synthetic 3X ETF (the risk-free asset) next month. The synthetic 3X ETF earns three times the monthly returns of the underlying 1X ETF.

They also consider a more realistic test using SPDR S&P 500 ETF (SPY) as the underlying 1X ETF and Direxion Daily S&P 500 Bull 3X Shares (SPXL) as the associated leveraged ETF. They assume 0.2% trading frictions for portfolio turnover. Using daily returns for 76 leveraged equity ETFs matched to 30 underlying 1X ETFs and daily implied volatilities for associated ATM call and put options during January 2007 through December 2021, they find that: Keep Reading

Can holders of popular large-capitalization stocks improve portfolio performance by systematically buying or selling options on these stocks? In their February 2024 paper entitled “The Performance of Options-Based Investment Strategies: Evidence for Individual Stocks from 2004 to 2019”, Zhuo Li and Thomas Miller, Jr. compare to buy-and-hold the performances of four strategies that augment a long stock position with options, as follows:

1. Buy and hold the stock.
2. Covered call  – long stock plus short call.
3. Protective put – long stock plus long put.
4. Collar – long stock plus short call plus long put.
5. Covered combination – long stock plus short call plus short put.

They focus on 10 stocks widely held in 401(k) plans: ExxonMobil, Comcast, Berkshire Hathaway (Class B), Oracle, Microsoft, Coca-Cola, Amazon, Wells Fargo, Google (Class A) and Apple. They roll at the end of each calendar month from the standard monthly option that expires during the next month to the one that expires during the subsequent month. They choose option strike prices that are at least 5% out-of-the-money but as close to 5% as possible, with exceptions when no such options are available. They assume option buys and sells are at the daily closing bid-ask midpoint. They ignore the possibility of early option exercise. Using monthly data for the selected 10 stocks and specified options as available during January 2004 through November 2019, they find that: Keep Reading