Economic Indicators

“Chicago Fed NFCI as U.S. Stock Market Predictor” suggests that weekly change in the Federal Reserve Bank of Chicago’s National Financial Conditions Index (NFCI) may be a useful indicator of future U.S. stock market returns. We test its practical value via two strategies that are each week in SPDR S&P 500 (SPY) when prior change in NFCI is favorable and in cash (U.S. Treasury bills, T-bills) when prior change in NFCI is unfavorable, as follows:

1. Change in NFCI < Mean [aggressive]: hold SPY (cash) when prior-week change in NFCI is below (above) its mean since since the beginning of 1973, providing an initial 20-year calculation interval.
2. Change in NFCI < Mean+SD [conservative]: hold SPY (cash) when prior-week change in NFCI is below (above) its mean plus one standard deviation of weekly changes in NFCI since the beginning of 1973.

The return week is Wednesday open to Wednesday open (Thursday open when the market is not open on Wednesday) per the NFCI release schedule. We assume SPY-cash switching frictions are a constant 0.1% over the sample period. We use buying and holding SPY as the benchmark. Using weekly levels of NFCI since January 1973 and weekly dividend-adjusted opens of SPY and T-bills since February 1993 (limited by SPY), all through April 2020, we find that: Keep Reading

A subscriber suggested that the Federal Reserve Bank of Chicago’s National Financial Conditions Index (NFCI) may be a useful U.S. stock market predictor. NFCI “provides a comprehensive weekly update on U.S. financial conditions in money markets, debt and equity markets, and the traditional and ‘shadow’ banking systems.” It consists of 105 inputs, including the S&P 500 Implied Volatility Index (VIX) and Senior Loan Officer Survey results. Positive (negative) values indicate tight (loose) financial conditions, with degree measured in standard deviations from the mean. The Chicago Fed releases NFCI each week as of Friday on the following Wednesday at 8:30 a.m. ET (or Thursday if Wednesday is a holiday), renormalized such that the full series always has a mean of zero and a standard deviation of one (thereby each week changing past values, perhaps even changing their signs). To investigate its usefulness as a U.S. stock market predictor, we relate NFCI and changes in NFCI to future S&P 500 Index returns. Using weekly levels of NFCI and weekly closes of the S&P 500 Index during January 1971 (limited by NFCI) through April 2020, we find that: Keep Reading

A subscriber requested testing Martin Zweig’s Combined Super Model, which each month specifies an equity allocation based on a system that assigns up to eight points from his Monetary Model and 0 or 2 points from his Four Percent Model. We consider two versions of the Combined Super Model:

1. Zweig-Cash – Allocate to Fidelity Fund (FFIDX) as equities, with the balance in cash earning the 3-month U.S. Treasury bill (T-bill) yield.
2. Zweig-FGOVX – Allocate to FFIDX as equities, with the balance in Fidelity Government Income Fund (FGOVX)

The benchmark is buying and holding FFIDX. We focus on compound annual growth rate (CAGR), maximum drawdown (MaxDD) and annual Sharpe ratio, with average monthly T-bill yield during a year as the risk-free rate for that year. We ignore impediments to mutual fund trading and any issues regarding timeliness of allocation changes for end-of-month rebalancing. Using monthly Combined Super Model allocations and monthly fund returns/T-bill yield during December 1986 through March 2020, we find that: Keep Reading

Why does the coincident relationship between U.S. aggregate corporate earnings growth and stock market return change from negative in older research to positive in recent research? In their January 2020 paper entitled “Assessing the Structural Change in the Aggregate Earnings-Returns Relation”, Asher Curtis, Chang‐Jin Kim and Hyung Il Oh examine when the change in the aggregate earnings growth-market returns relationship occurs. They then examine factors explaining the change based on asset pricing theory (expected cash flow and expected discount rate). They calculate aggregate earnings growth as the value-weighted average of year-over-year change in firm quarterly earnings scaled by beginning-of-quarter stock price. They consider only U.S. firms with accounting years ending in March, June, September or December, and they exclude firms with stock prices less than $1 and firms in the top and bottom 0.5% of quarterly earnings growth. They calculate corresponding quarterly stock market returns from one month prior to two months after fiscal quarter ends to capture earnings announcement effects. Using quarterly earnings and returns data as specified for a broad sample of U.S. public firms from the first quarter of 1970 through the fourth quarter of 2016, they find that:

Keep Reading

Do investors drive stocks to overvaluation (undervaluation) in good (bad) economic times, such that corresponding expectations for future returns are therefore relatively low (high). In the August 2019 update of their paper entitled “Consumption Fluctuations and Expected Returns”, flagged by a subscriber, Victoria Atanasov, Stig Møller and Richard Priestley introduce the cyclical consumption economic variable and examine its power to predict stock market returns. They hypothesize that in good (bad) economic times:

1. Marginal utility of present consumption is low (high).
2. Investors are willing (unwilling) to sacrifice current consumption for investment.
3. This investment pushes stock prices up (down) and expected returns therefore down (up).

Their principal measure of consumption is quarterly seasonally adjusted real per capita consumption expenditures on non-durables and services from the National Income and Product Accounts (NIPA) Table 7.1 maintained by the U.S. Bureau of Economic Analysis. They extract its cyclical component (detrend) by regressing the logarithm of real per capita consumption on a constant and four lagged values of consumption from about six years prior. They conduct both in-sample and out-of-sample (expanding window regressions, with 2-quarter lag for release delay) tests of the quarterly relationship between cyclical consumption and future U.S. stock market returns. Using the specified consumption data and quarterly returns for the S&P 500 Index and the broad value-weighted U.S. stock market from the first quarter of 1947 through the fourth quarter of 2017, they find that: Keep Reading

A subscriber requested review of an analysis concluding that combining economic trend and market trend signals enhances market timing performance. Specifically, per the example in the referenced analysis, we look at combining:

• The 10-month simple moving average (SMA10) for the broad U.S. stock market. The trend is positive (negative) when the market is above (below) its SMA10.
• The 12-month simple moving average (SMA12) for the U.S. unemployment rate (UR). The trend is positive (negative) when UR is below (above) its SMA12.

We consider scenarios when the stock market trend is positive, the UR trend is positive, either trend is positive or both trends are positive. We consider two samples: (1) dividend-adjusted SPDR S&P 500 (SPY) since inception at the end of January 1993 (nearly 26 years); and, (2) the S&P 500 Index (SP500) since January 1948 (limited by UR availability), adjusted monthly by estimated dividends from the Shiller dataset, for longer-term robustness tests (nearly 71 years). Per the referenced analysis, we use the seasonally adjusted civilian UR, which comes ultimately from the Bureau of Labor Statistics (BLS). BLS generally releases UR monthly within a few days after the end of the measured month. We make the simplifying assumptions that UR for a given month is available for SMA12 calculation and signal execution at the market close for that same month. When not in the stock market, we assume return on cash from the broker is the yield on 3-month U.S. Treasury bills (T-bill). We focus on gross compound annual growth rate (CAGR), maximum drawdown (MaxDD) and annual Sharpe ratio as key performance metrics. We use the average monthly T-bill yield during a year as the risk-free rate for that year in Sharpe ratio calculations. While we do not apply any stocks-cash switching frictions or tax considerations, we do calculate the number of switches for each scenario. Using specified monthly data through September 2019, we find that: Keep Reading

A reader requested review of the Decision Moose asset allocation framework. Decision Moose is “an automated stock, bond, and gold momentum model developed in 1989. Index Moose uses technical analysis and exchange traded index funds (ETFs) to track global investment flows in the Americas, Europe and Asia, and to generate a market timing signal.” The trading system allocates 100% of funds to the index projected to perform best. The site includes a history of switch recommendations since the end of August 1996, with gross performance. To evaluate Decision Moose, we assume that switches and associated trading returns are as described (out of sample, not backtested) and compare the returns to those for dividend-adjusted SPDR S&P 500 (SPY) over the same intervals. Using Decision Moose signals/performance data and contemporaneous SPY prices during 8/30/96 through 9/30/19 (23+ years), we find that: Keep Reading

How do different asset classes interact with the Japanese yen-U.S. dollar exchange rate? To investigate, we consider relationships between Invesco CurrencyShares Japanese Yen (FXY) and the exchange-traded fund (ETF) asset class proxies used in “Simple Asset Class ETF Momentum Strategy” (SACEMS) at a monthly measurement frequency. Using monthly dividend-adjusted closing prices for FXY and the asset class proxies since March 2007 as available through July 2019, we find that: Keep Reading

In response to “OFR FSI as Stock Market Return Predictor”, a subscriber suggested overlaying a 10-month simple moving average (SMA10) technical indicator on the Office of Financial Research Financial Stress Index (OFR FSI) fundamental indicator for timing SPDR S&P 500 (SPY). The intent of the suggested overlay is to expand risk-on opportunities safely. To test the overlay, we add four strategies (4 through 7) to the prior three, each evaluated since January 2000 and since January 2009:

1. SPY – buy and hold SPY.
2. OFR FSI-Cash – hold SPY (cash as proxied by 3-month U.S. Treasury bills) when OFR FSI at the end of the prior month is negative or zero (positive).
3. OFR-FSI-VFITX – hold SPY (Vanguard Intermediate-Term Treasury Fund Investor Shares, VFITX, as a more aggressive risk-off asset than cash) when OFR FSI at the end of the prior month is negative or zero (positive).
4. SMA10-Cash – hold SPY (cash) when the S&P 500 Index is above (at or below) its SMA10 at the end of the prior month.
5. SMA10-VFITX – hold SPY (VFITX) when the S&P 500 Index is above (at or below) its SMA10 at the end of the prior month.
6. OFR-FSI-SMA10-Cash – hold SPY (cash) when either signal 2 or signal 4 specifies SPY. Otherwise, hold cash.
7. OFR-FSI-SMA10-VFITX – hold SPY (cash) when either signal 3 or signal 5 specifies SPY. Otherwise, hold VFITX.

Using end-of-month values of OFR FSI, SPY total return and level of the S&P 500 Index during January 2000 (OFR FSI inception) through June 2019, we find that:

Keep Reading

Is the Office of Financial Research Financial Stress Index (OFR FSI), described in “The OFR Financial Stress Index”, useful as a U.S. stock market return predictor? OFR FSI is a daily snapshot of global financial market stress, distilling more than 30 indicators via a dynamic weighting scheme. The index drops and adds indicators over time as some become obsolete and new ones become available. Unlike some other financial stress indicators, past OFR FSI series values do not change due to any periodic renormalization and are therefore suitable for backtesting. To investigate OFR FSI power to predict U.S. stock market returns, we relate level of and change in OFR FSI to SPDR S&P 500 (SPY) returns. Using daily and monthly values of OFR FSI and SPY total returns during January 2000 (OFR FSI inception) through June 2019, we find that:

Keep Reading