{"id":204618,"date":"2024-04-08T11:35:32","date_gmt":"2024-04-08T15:35:32","guid":{"rendered":"https:\/\/ibkrcampus.com\/?p=204618"},"modified":"2024-09-11T13:30:10","modified_gmt":"2024-09-11T17:30:10","slug":"economic-momentum","status":"publish","type":"post","link":"https:\/\/www.interactivebrokers.com\/campus\/ibkr-quant-news\/economic-momentum\/","title":{"rendered":"Economic Momentum"},"content":{"rendered":"\n

The article \u201cEconomic Momentum\u201d first appeared on Alpha Architect blog<\/a>.<\/em><\/p>\n\n\n\n

Out of the hundreds of exhibits in the factor zoo, momentum was one of just five equity factors that met all the criteria (persistent, pervasive, robust, implementable, and intuitive) Andrew Berkin and I established in our book Your Complete Guide to Factor-Based Investing<\/em><\/a>. Because of the strong empirical evidence, momentum\u2014the tendency of assets that have performed well recently (e.g., over the prior year) to outperform assets in the same asset class that has performed poorly over the same period\u2014continues to receive much attention from researchers.<\/p>\n\n\n\n

While the original 1993 research<\/a> on momentum by Narasimhan Jegadeesh and Sheridan Titman focused solely on U.S. common stock returns, momentum\u2014both cross-sectional (which looks at the relative strength of a cross-section of markets, going long the assets with the most relative strength and short the assets with the least relative strength) and time-series, or trend (which goes long the assets that have been rising and short the assets that have been falling)\u2014has been found pretty much everywhere<\/a>: in global stocks, government and corporate bonds, commodities, currencies,  stock portfolios, industries, countries, and even options<\/a>.<\/p>\n\n\n\n

Adding to the momentum literature, in his February 2015 NBER paper \u201cFundamentally, Momentum is Fundamental Momentum,<\/a>\u201d Robert Novy-Marx presented evidence demonstrating that momentum in stock prices isn\u2019t an independent anomaly. Instead, it\u2019s driven by fundamental momentum. Marx found that momentum in firm fundamentals (i.e., earnings momentum) explains the performance of strategies based on price momentum. It holds for both large and small stocks. Marx\u2019s findings on fundamental momentum were consistent with the findings of Shuoyuan He and Gans Narayanamoorthy, authors of the September 2017 study \u201cEarnings Acceleration and Stock Returns<\/a>,\u201d who found that earnings acceleration is a significant predictor of future stock returns\u2014going long in the top decile of quarterly earnings acceleration and short in the bottom decile of earnings acceleration produced large market-adjusted returns.<\/p>\n\n\n\n

Based on these findings, at least one investment firm, AQR Capital, has incorporated fundamental momentum as well as price momentum (both cross-sectional and time-series), into their fund construction strategies. As measures of fundamental momentum, AQR uses three fundamental metrics: earnings momentum, analysts\u2019 revisions, and margin growth.<\/p>\n\n\n\n

The findings raise an interesting question: Could fundamental momentum exist elsewhere? Because investors underreact to fundamental<\/em> stock information, perhaps they also underreact to fundamental <\/em>economic information.<\/p>\n\n\n\n

Latest Research<\/strong><\/h2>\n\n\n\n

AQR\u2019s Jordan Brooks, Noah Feilbogen, Yao Hua Ooi, and Adam Akant, authors of the February 2024 study \u201cEconomic Trend<\/a>,\u201d examined whether there was a tendency for new information to have a persistent impact on asset prices by positioning in each market based on trends in macroeconomic fundamentals. The authors built on the findings of Jordan Brooks\u2019 original 2017 study \u201cA Half Century of Macro Momentum<\/a>,\u201d which covered the period 1970-2016. His findings led Brooks to conclude:<\/p>\n\n\n\n

\n

\u201cMacro momentum, a systematic approach to global macro investing that takes long positions in assets with improving fundamental trends and short positions in assets with deteriorating fundamental trends, has performed consistently over nearly the last half century. The strategy is highly diversifying to traditional assets, tending to provide an excellent hedge in both large equity market drawdowns and rising real yield environments.\u201d<\/p>\n<\/blockquote>\n\n\n\n

The authors extended the original strategy described in Brooks\u2019 2017 paper and focused solely on its directional implementation. To add an out-of-sample test that covered challenging economic times, \u201cthe equity market sell-off of 2018, the global pandemic, and the subsequent emergence of global inflation<\/a> and associated monetary policy response,\u201d they extended the data through 2022. They referred to the \u201cresulting strategy as \u2018economic trend,\u2019 an acknowledgment of the similar underlying economic mechanism and attractive performance characteristics it shares with price trend-following.\u201d They began by noting:<\/p>\n\n\n\n

\n

 \u201cEconomic trend and price trend stem from the same premise: asset prices tend to underreact to new information, which, for macro assets, typically takes the form of news about economic fundamentals. In other words, when fundamentals shift, markets adjust gradually. This gradual incorporation of news creates trends in asset prices.\u201d While price trend is a simple strategy, economic trend \u201crequires identifying the relevant fundamental variables that drive each asset\u2019s price (e.g., inflation expectations impact bond prices), modeling trends in these fundamental variables in real-time (e.g., measuring real-time trends in inflation expectations for each market), and a structural mapping from fundamental trends to asset prices (e.g., increasing inflation expectations in a country is bad news for its stock and bond markets, good news for its currency, etc.). Unlike price trend, economic trend will not invest where price action isn\u2019t supported by underlying fundamental news. The strategy will suffer when returns are disconnected from changes in fundamentals (as can happen) but will generate profits when asset prices begin to cue off fundamental news again.\u201d<\/p>\n<\/blockquote>\n\n\n\n

Methodology<\/strong><\/h2>\n\n\n\n

For both growth and inflation, the authors used 12-month changes in GDP growth forecasts and CPI inflation forecasts. For international trade, they used 12-month changes in the local FX spot rate<\/a> against an export-weighted basket. They measured monetary policy trends using 12-month changes in short-term bond yields. And they measured risk aversion trends using an equal-weighted combination of 12-month trailing equity market returns and 12-month changes in credit spreads.<\/p>\n\n\n\n

The authors explained:<\/p>\n\n\n\n

\n

\u201cThe signs of these relationships are based on theoretically motivated and empirically verified \u2018other things equal\u2019 contemporaneous relationships between returns and news about each theme. Economic trend is built on the premise that news tends to have a persistent impact on asset prices, which implies that contemporaneous and predictive relationships between returns and news must have the same sign. This restriction reduces the scope for data mining.\u201d As an example, \u201cIncreasing growth is good news for equity markets. Likewise, strong equity market performance is associated with declining inflation, expansionary monetary policy trends, improving trade competitiveness, and declining risk aversion.\u201d<\/p>\n<\/blockquote>\n\n\n\n

Their asset universe included 15 global equity futures, nine bond futures, seven interest rate futures, eight currency forwards, and 20 commodity futures.  For each of the five economic themes, they created a portfolio for each of the four asset classes, leaving us with 20 asset class theme portfolios. The table below shows the signs of each of the relationships:<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

They then aggregated these asset class theme portfolios into equal risk-weighted composite portfolios, forming three types of composite portfolios:<\/p>\n\n\n\n

1) Asset class composite portfolios. These combine all theme views at equal risk weight within a given asset class. For example, the equity asset class composite portfolio invests in equity index futures using growth, inflation, international trade, monetary policy, and risk aversion as inputs.<\/p>\n\n\n\n

2) Theme composite portfolios. These trade a single theme across all asset classes. For example, the growth theme composite portfolio invests in equity index futures, bond futures, interest rate futures, currency forwards, and commodity futures, using views based solely on growth.<\/p>\n\n\n\n

3) Aggregate economic trend composite portfolio. This is an equal risk-weighted composite across all asset class theme portfolios. It represents the full economic trend strategy, implemented using all themes and assets.<\/p>\n\n\n\n

For each asset class, they formed analogous portfolios for price trends, using an equal-weighted combination of one-month, three-month, and 12-month trailing asset returns as factors. The price trend asset class and aggregate composite portfolios used the same universe and portfolio construction described above.  All composite portfolios (both economic trend and price trend) targeted 10% constant volatility at each point in time using a trailing three-year covariance matrix.<\/p>\n\n\n\n

Following is a summary of their key findings:<\/p>\n\n\n\n