Donald Trump’s campaign to “Make America Great Again” is known for winning — and not just at the polls. Trump routinely says things like “We will win again. We will win a lot,” and this rhetorical tic has been fodder for pundits and comedians alike.

But if you look closely, Trump talks far more about losing than he does about winning. It’s this “losing” language that could be the key to his success to date. My research shows that making people, and especially Republicans, think about losses motivates them to take political action.

I analyzed all the primary debates through March 6, counting how many times each candidate talked about losing (words like “loss,” “lose,” and “don’t win”). Trump talks about losing two to five times more than other candidates — once every 350 words or so.

This means that if the debate transcripts were printed on double-spaced pages, Trump’s words would take up about 100 pages, and on average he would mention losing on at least two of every three of those pages.

The way Trump talks about losses is also unique. He is the only candidate who primarily talks about loss in the context of jobs, the economy, or losing in general.  Other candidates mostly mention losing in the context of armed conflict or the current elections.

Furthermore, Trump is much more likely to talk about present or future losses, rather than how the United States has lost in the past. This matters because instead of treating losses as “sunk costs” — past costs from which we can and should move on — Trump generates additional fear and anxiety by emphasizing the looming prospect of ongoing loss now and in the future.
Why does this matter? People dislike losses much more than they like gains. When potential losses are coupled with elevated anxiety, people pay attention more to issues and are more likely to be persuaded by political arguments.

An experiment I conducted shows how this works. I told 1,000 people that the U.S. government was considering an increase in the federal gas tax. Later on, everyone had the chance to sign an online petition supporting or opposing the new policy (they could choose either option).  I measured how many people decided to sign.

The twist was that half the participants first were told about the possible costs for the proposed policy — things like paying an extra $3.50 to $5 each  time a person fills their tank and the higher cost of food and other goods.

The other participants were told about the possible benefits of the policy — things like more funds to improve declining infrastructure, road and bridge repairs, and better public transit. All participants were asked to imagine and write about how the policy would affect people like them.

Then, before offering people the petions,  I measured how focused they were using what psychologists call a Stroop test. Participants were shown words like “BLUE” and “YELLOW” in different colors. Sometimes the word color and meaning matched, but other times the words were printed in other colors, such as red or green. The goal is to enter the color of the word, not the color spelled by the word). But the brain finds it confusing when word and color do not match and requires extra focus to get the answer correct.

I find that when people are told about the costs of a policy, their level of focus is higher afterward, regardless of their political ideology. But this increased attention is far more consequential for motivating conservatives to take action.

Whether they saw the costs or benefits, about 22 percent of liberals and 20 percent of moderates signed the online petition.

For conservatives, however, the costs-benefits framing mattered a lot. Less than 9 percent of conservatives signed the petition when shown policy benefits. However, when conservatives saw policy costs, their participation rate jumped threefold to over 26 percent. Like Republicans listening to Trump’s speeches, conservatives in my study took action when told how they would lose under future policies.

This finding matches other evidence. Negative images and messages capture the attention of conservatives much more effectively than of liberals. For example, conservatives find and look longer at angry and scary images. When faced with risk, conservatives show greater activity in the amygdala — a key area in the brain linked to threat detection — as compared with liberals.

This finding also matches some evidence from the Republican primary, where turnout is expected to be at least six points higher than any other primary in the past 36 years.

Of course, this is just one study of one policy issue, and it is not focused on Trump. But the results suggest why Trump is so effective at getting Republicans to vote (and to vote for him): He evokes a fear of economic losses. And because Republicans are particularly responsive to the specter of loss, they are more likely to mobilize to protect what they have.

This is especially true, as an earlier Monkey Cage post noted, because people faced with future losses tend to make riskier decisions to try to avoid those losses. In this case, Trump is the risky choice — a Hail Mary pass intended to avert economic stagnation for middle-class America.

Elaine Denny is a doctoral candidate in political science and international relations at the University of California in San Diego.