Written by Krisna Patel, CFA, and edited by Hanna Jackson
Do you ever joke about the accuracy of weather forecasts? Honestly, I often dismiss any forecasts more than two days out and doubt that meteorologists can accurately predict anything beyond the next 24 hours. According to the National Oceanic and Atmospheric Administration, however, weather forecasts are accurate 90% of the time in a five-day forecast and a surprising 80% in a seven-day forecast[i]. Many of us may question this statistic because we naturally recall when the forecast was wrong, not when it was correct. This misperception stems from a notion called the availability bias. To make decisions as rapidly as possible, our brains take “mental shortcuts” by recalling the most readily available memories that led to the desired outcome. In the case of the weather, we typically recall the times our plans were changed or ruined due to forecast inaccuracies. As a result, we tend to largely mistrust those forecasts.
Quite the opposite is true for stock return forecasts, though. As a financial advisor, clients often ask me to predict the next month, quarterly, or even annual return, hoping to hear some magic answer. Each time, I hesitate to answer and quickly turn the conversation toward valuation and long-term capital market expectations. This, of course, doesn’t answer their question, but I aim instead to show them that I do not have a crystal ball. Large investment firms, on the other hand, often embrace the question, publicly posting their year-end price targets for the S&P 500®. I speak with managers at many of these firms to understand how they came to those conclusions and what factors affect their predictions. I have consistently found that their pronouncements of forecasted performance are not always accurate.
In July 2018, one would have been hard-pressed to find an investment house calling for a higher allocation to fixed income securities as opposed to equities over the coming year. According to Morningstar®, the actual realized return from July 2018 through June 2019 for the bond index proxy iShares® Core US Aggregate Bond ETF (AGG) was 7.72%, while the equity index proxy iShares® MSCI ACWI ETF (ACWI) returned only 5.99% with significantly greater volatility[ii]. Standard financial theory would consider this an anomaly because the investment with lower volatility provided a greater return—certainly not an outcome that analysts predicted.
Like the weather, investment firms’ forecasts might not be considerably reliable, but that does not mean that their predictions are not worth considering. Maybe they simply needed to be taken with a grain of salt. Michael Batnick, CFA, posted on his blog The Irrelevant Investor regarding this topic late last year. He compiled firms’ price targets going back to 2005 and found that, on average, the predicted return for the stock market sat at 9% on an annual basis, quite close to the average annualized return. Beyond that, though, he calculated that the average annual forecast was off by 10.7% over those thirteen years[iii]. That’s quite the miss.
Hassan, Pouran, and Reza Espahbodi—three accounting experts and retired professors—sought to determine the margin of error in analysts’ earnings predictions, using data from 1994 to 2013. They calculated that analysts were inaccurate in their nine-month forecasts by an average of 33.7%[iv]. That level of deviation from projected earnings would likely be make-or-break for a company’s stock price. Oddly enough, the research also found that these forecasts were increasingly off the mark after the implementation of government regulations intended to make predictions more accurate.
A 2018 study by the International Monetary Fund (IMF) focused on determining how well analysts predict recessions. They evaluated projections for 63 countries (29 advanced and 34 emerging) from 1992 to 2014. They found that of the 153 observed recessions only 5 were predicted 9 months prior by consensus forecasts. Three months prior to a recession consensus forecasts still missed 23% of the coming recession. The IMF economists didn’t fare much better, missing 147 of the 153 recessions nine months prior to a recession[v].
To distinguish between the two studies, I must note that neither study focused on price targets specifically. The Espahbodi study examined earnings forecasting accuracy, and the IMF studied GDP changes. At the same time, though, both metrics are often extensively used to predict the equity price target.
Because a focus on nailing the target seems futile, let’s widen the range of acceptable predictions. In 2013, Itzhak Ben-David, John Graham, and Campbell Harvey examined over 13,000 predictions made over ten years (1994-2013) primarily by CFOs of mid- to large-sized companies. They found that only 36% of the annual forecasted returns laid within the CFO’s 80% confidence interval for the S&P 500®[vi]. Their more recent research, addressed in their working paper The Persistence of Miscalibration, updated their 2013 numbers by including additional data up to third quarter of 2017. The numbers had only gotten worse: CFOs’ forecasted returns hit only 30.8% of the time in their 80% confidence interval[vii]. In their 2013 study, they also found that while the historical return dispersion of the S&P 500® between the 10th and 90th percentile is 42.2%, the CFOs’ confidence interval dispersion was only 14.5%. Ben-David, Graham, and Harvey surmised that overconfidence bias—belief that one’s given ability is better than the average person’s—was a primary culprit of these CFOs’ narrow interval of forecasted return.
In the end, answers to that dreaded question of the stock market’s direction should not be a prime indicator of potential performance because the short-term movement is overwhelmingly unstable. Human sentiment in the markets can suddenly change with a simple tweet, headline, or unforeseen geopolitical event. This sentiment is usually impossible to predict months in advance and can lead to markets being overbought or oversold for longer periods than fundamental predictors would otherwise suggest. Weather, on the other hand, is not vulnerable to superficial variables like human sentiment but is instead dictated and limited by scientific and physical fact. Meteorologists can say with 100% certainty, for example, that there will be no snow in 80-degree weather due to scientific impossibility. Unfortunately, when it comes to stock forecasts, there are no definable parameters on future outcomes.
So, tell me again, what will the weather look like next week? Dismissive or not, I will probably believe that forecast more readily than the market predictions.
Edited by Hanna Jackson, registered assistant to Krisna Patel
Past performance is not a guarantee of future results. Information presented herein is for discussion and illustrative purposes only and is not a recommendation or an offer or solicitation to buy or sell any securities. Views expressed are as of 07/15/2019, based on the information available at that time, and may change based on market and other conditions. Although certain information has been obtained from sources believed to be reliable, we do not guarantee its accuracy, completeness or fairness. We have relied upon and assumed without independent verification, the accuracy and completeness of all information available from public sources.
Krisna Patel is an Investment Advisor Representative at Engage Financial Group--11622 North Michigan Road, Zionsville, IN 46077.
Securities and investment advisory services offered through Woodbury Financial Services, Inc. (WFS), member FINRA/SIPC. WFS is separately owned and other entities and/or marketing names, products or services referenced here are independent of WFS.
[i] Source: NOAA SciJinks. Published on https://scijinks.gov/forecast-reliability/.
[ii] Source: Morningstar® Advisor Workstation. Data as of 07/10/2019.
[iii] Michael Batnick, The Irrelevant Investor, “Next Year.” Published on https://theirrelevantinvestor.com/ 12/17/2018.
[iv] Hassan Espahbodi, Pouran Espahbodi, Reza Espahbodi, “Did Analyst Forecast Accuracy and Dispersion Improve after 2002 Following the Increase in Regulation?” Published in the Financial Analysts Journal®, Volume 71, Number 5, September/October 2015.
[v] Zidong An, João Tovar Jalles, Prakash Loungani, “How Well Do Economists Forecast Recessions?” Published on https://www.imf.org/ 03/05/2018.
[vi] Itzhak Ben-David, John R. Graham, Campbell R. Harvey, “Managerial Miscalibration.” Published in The Quarterly Journal of Economics 08/13/2013.
[vii] Itzhak Ben-David, John R. Graham, Campbell R. Harvey, “The Persistence of Miscalibration” Working Paper.