The whole world is waiting for the answer. A vaccine is the key is to getting economies back to normal and potentially saving millions of lives.

Governments around the world have committed billions to help bring a vaccine to market as soon as one proves to be safe and effective in human trials.

A vaccine from the University of Oxford, in the U.K., which is under development in partnership with the pharmaceutical company AstraZeneca, is already in large-scale clinical trials, with results expected sometime in August.

The drugmaker has said if the vaccine works, and if regulators approve it on an emergency basis, it will be able to begin delivering hundreds of millions of doses by the end of September.

So, you might think the chances of having a vaccine this year are pretty good. Besides Oxford’s vaccine, there are at least 162 different vaccine candidates under development, with at least 22 in some stage of clinical trials, according to data published this week by the World Health Organization.

A lot of companies have certainly baked the prospect of a viable vaccine coming to market this year into their forecasts. For instance, Pictet Wealth Management, one of Europe’s best-known asset managers, told its clients on a conference call last month that its central forecast put the probability of a vaccine by year-end at 80%, according to a Bloomberg News report.

But there’s at least one good reason to think that, this time, the smart money isn’t so smart.

Wisdom of crowds

Prediction markets—exchanges in which participants buy and sell contracts on future events in fields from politics and entertainment to foreign affairs and science—are indicating there will not be a COVID-19 vaccine this year.

Research has shown that these specialized markets, which are primarily designed to aggregate “the wisdom of crowds,” rather than allocate capital, produce more accurate predictions than either individual experts or public opinion polls. Companies even use them internally to help forecast whether large projects will be completed on time and within budget, or to forecast future sales.

On issues of public health, past studies have, for example, shown that prediction markets can more accurately forecast seasonal flu outbreaks between than other modeling tools.

And, right now, the prediction markets are betting firmly against a vaccine being available before January.

The French company Hypermind runs one of the largest public prediction markets. It currently puts the chances of the U.S. Food and Drug Administration approving a coronavirus vaccine this year at less than one in three. In fact, the market indicates that the best chance—59% likelihood—is that a vaccine will only be approved sometime after the first quarter of 2021.

Prediki.com, another online play-money prediction market, is even more pessimistic. It puts the chances of a vaccine being mass-produced before January at one in five. In fact, the odds of a vaccine only cross 50-50 after June 2021.

Several other prediction markets—including the real-money Iowa Electronic Markets which trades U.S. political futures contracts—have not offered a contract on a COVID-19 vaccine.

Researchers at Singapore’s National University are currently running an experiment to see if prediction markets produce better forecasts than individual opinions about the course of the pandemic, according to a notification listed on the U.S. National Library of Medicine’s clinical trials page. But results from this experiment are not yet available.

Participants in Hypermind’s market use play money to buy and sell futures contracts, to avoid running afoul of financial securities laws. But Hypermind does award real money prizes to those whose forecasts come true.

Emile Servan-Schreiber, Hypermind’s managing director, says that the firm’s own research indicates trading in prediction markets arrive at a forecast of events that is more accurate than random chance very early on after trading in a contract begins. On average, however, the contract with the best odds accurately aligns the actual outcome of events after about a third of a contract’s life.

Hypermind’s COVID-19 vaccine contract began trading in May—so it hasn’t yet run for a third of its contract life—and has had between 100 and 120 buyers and sellers so far, Servan-Schreiber said.

And, in sign that not all hope for a vaccine this year is lost, in the past two days, there has been a flurry of trading in the market on news that Moderna’s vaccine candidate produced antibodies to the coronavirus in early patient tests, he said. The Moderna report helped send stock markets higher in Europe and the U.S. on Wednesday on hopes of a vaccine breakthrough in the pipeline. And, Moderna shares rocketed 17% higher at the open.

The odds of a vaccine being delivered this year have tightened significantly: now slightly less than one-in-three, they used to be less than one-in-five.

Hypermind has much better news for the global airline industry: it is also taking trades on the question of when global commercial airline traffic will rise above 75,000 flights per day. (This figure fell below 25,000 daily flights in early April and is currently at about 61,000 flight per day, according to data from flight tracking website FlightRadar24.) Right now the contract puts a 63% chance of that happening in the third quarter and a 91% chance of it happening by year end.