Breakingviews - Stricken polls can work to investors’ advantage

"I Voted" stickers are seen during early voting at the Brooklyn Museum in the Brooklyn borough of New York City, New York, U.S. October 29, 2020. REUTERS/Brendan McDermid

LONDON (Reuters Breakingviews) - Investors can’t be blamed for wanting to swear off polls. Once again, U.S. opinion surveys underestimated Republican support in last week’s presidential election. The trouble is investors don’t have much of an alternative. With two cycles showing clear – and now known – biases, perhaps investors just need to use erroneous polls to their advantage.

Though ballots are still being counted, it doesn’t look great for the industry. Polls were off by an average of 6 percentage points in the 39 states where 98% or more of votes are already in, according to a Breakingviews analysis using FiveThirtyEight polling averages. In all but one state, Louisiana, they understated Republican Donald Trump’s vote share relative to victorious Democratic Party candidate Joe Biden.

The knee-jerk reaction of forswearing polls, however, would leave election-watchers worse off. Prediction markets, which allow punters to lay wagers on votes, are the obvious backup. Investors can see the implied probability of outcomes based on the prices. But relatively thin liquidity and high transaction costs make them less efficient than financial markets. A 2012 study in the Electoral Studies journal found that they mostly just incorporated polling information, rooting them in the same potentially erroneous information.

Then there are the so-called economy-based models, which try to derive the likely winner using metrics like unemployment and GDP growth. It makes intuitive sense that voters would favour an incumbent party while the economy is roaring. Yet a famous model from Yale University economist Ray Fair has clocked up some large errors in recent years. One problem is that voters’ political perspective often colours how they view the economy, rather than the other way around.

The upshot is that forecasting is hard, and surveys which simply ask people how they will vote may still be the best guide. So if investors can keep the ever-present danger of polling errors front of mind, they may even make some money. After all, Dan Loeb’s Third Point gained nearly $400 million by making the right call on the post-election market, according to the Financial Times. A diligent reading of the polls this year showed that Biden’s lead was big enough to withstand a large error. As with any trade, it isn’t the data that is presented, but how investors interpret it that matters.


Reuters Breakingviews is the world's leading source of agenda-setting financial insight. As the Reuters brand for financial commentary, we dissect the big business and economic stories as they break around the world every day. A global team of about 30 correspondents in New York, London, Hong Kong and other major cities provides expert analysis in real time.

Sign up for a free trial of our full service at and follow us on Twitter @Breakingviews and at All opinions expressed are those of the authors.