Why You Should Never Loan Money to - or Vote for - Someone Who Invokes God

I just finished Seth Stephens-Davidowitz's fascinating book, Everybody Lies. One of the ideas behind this is that we are more likely to reveal ourselves in private searches than we are in social media posts. Interestingly, there is data on what Americans look for when they search and thus, data on who we are.

One of the things he shares is words that predict whether someone is more or less likely to pay back a loan. A website prosper lets people offer a story about why they need a loan and folks reading this can decide whether or not to loan them money. Three economists analyzed the stories and tried to identify which words would be a leading indicator of who would pay back loans and who would default. (About 13% of people who borrow do default, never paying back the loan.) This is what they found.

Terms used in loan applications by people most likely to pay back:
debt-free, after-tax, graduate, lower interest rate, minimum payment

Terms used in loan applications by people most likely to default:
God, promise, will pay, hospital, thank you

As it turns out, people who are dealing with reality are more likely to pay back. These people are talking about real issues, like after-tax income, or minimum payments. It also makes sense that someone mentioning when they might graduate would be a good risk; this person should soon see a drop in tuition and - hopefully - a rise in salary. Their situation is about to change for the better and it makes sense that they would be in better position tomorrow than they are today; they're a good risk.

Meanwhile, people who offer only their word (promise or will pay) aren't forecasting any change in their future situation and thus there is little reason to believe that they'll soon go from needing money to having extra money and being able to pay you back. Hospital, too, suggests why they would need money now but really offers no reason why they'd have more money in the future. (The mention of something as expensive as hospital is an argument for healthcare, not a loan.) Finally, there is good reason why mention of God should be a red flag.

People who mention God in their requests for loans are 2.2X more likely to default than the average person. This is the highest-risk word and it makes perfect sense why that would be.

Only two people would mention God in regards to a loan: people who think he cares about their financial situation and people who are trying to con the religious.

People who think that God cares about their finances cannot, of course, explain why God would first get cash to someone in a first-world country before he would a small child in a third-world country. It's a ludicrous idea on so many fronts and the God who would care that you got a new car that let you stop riding the bus while a child in a refugee camp goes without education is an odd God. Further, these people are not dealing with reality. They're not talking about or thinking about minimum payments or graduation. They are waiting for God to intervene in their finances. That's not likely to happen.

The other folks who would invoke God when asking for money are running a con. Religious people are conscientious and easier to manipulate once you've reminded them of their faith. The mention of God by these con men is done because it is a more effective way to get your money. If they thought that mention of the San Diego Padres was more likely to prompt you to give them money, they'd wax eloquent about the Padres.

The same arguments hold for someone asking for your vote rather than a loan. Anyone who invokes God is either hoping that God will intervene in policy matters or running a con on voters who believe that. Given a choice between a plumber who has training and experience and one who believes the support of the evangelical community will make him successful when he crawls under your kitchen sink for the first time, I'll go with the plumber who has demonstrable skills and - regardless of what God or afterlife he believes in - believes that natural consequences comes from natural - not supernatural - intervention. I'd rather have a plumber with a good wrench than one with a great prayer. And I don't see any evidence that policy is any different.

If someone mentions God when asking for a loan or asking for your vote, just say no. Instead, ask them how they are going to deal with reality. If they are making $1,500 a month now and need a loan for $500, what is going to change in a month when they begin paying back the loan that makes me think they'll have an extra $100 a month rather than be running short by $100 a month? God is not the right answer.

The question of any politician is not whether he thinks God likes him better than his opponent. The question of the guy who wants your vote is no different than the question of the guy who wants a loan: how are you going to deal with reality, making the rash assumption that no miracle will save you?

The Test for Gods and Worldviews

Once upon a time people had a fairly simple test for the power of a God. They simply wanted to know, did he help you to win battles? Constantine supposedly adopted Christianity before a major battle and then won it, cementing his conversion and prompting him to make Christianity the official religion of the Roman Empire.  When Romans proved themselves stronger with their god, other communities took this as proof that this was the god to worship. A few centuries later, a similar wave of "proof" in the form of power and conquest swept down the silk routes from northern Africa to China with Islam.

It is easy to mock this test but it's not the worst test of efficacy; does adopting your god make us more powerful, more able to get what we want? For whatever limits we have in making arguments or accepting data that contradicts deeply held beliefs, we might still benefit from this simple test: does that worldview make people prosperous or poor? Does it give you the power to live a life of your choosing?

Right now facts seem to have gone out of style. Republicans and their president have largely dismissed the importance of facts, particularly the ones that challenge their worldview. People worry that the party - the country even - will never recover and will be hijacked by irrational worldviews that weaken the country. It might.

I think, though, that this worldview is too ineffectual to last long.

In San Francisco, average household income in 2015 was $119,406. In West Virginia average income was less than half that, $56,425.  Less than half of San Francisco's population is white; 94% of West Virginians are white.

Making the assumption that every baby has the same potential at conception - regardless of race or gender - San Francisco is obviously doing a better job of creating prosperity for its residents. It might be that their residents are genetically superior to West Virginians who suffer from a lack of racial diversity; or it might be that the worldview of San Franciscans is more effective than that of West Virginians.

There are a variety of ways to define the difference in worldview but one simple one is to look at how these two communities voted. Fewer than 10% of San Franciscans voted for Trump; more than two-thirds of West Virginians did.

One of two things will happen in our country. Either the Trumpian worldview will spread like a virus and the country will become like West Virginia, a place where Trump holds rallies when he wants comfort. Or people in places like West Virginia will see the worldview of the people in San Francisco - a worldview that embraces diversity and disruption, a place that funds startups rather than tries to protect a coal industry that first emerged in 1740 - as more powerful and adopt it. If the first happens, the country will look back at the 20th century as a golden time; if the second happens, West Virginians will look back fondly at the time that their grandparents adopted a new worldview that made them prosperous and powerful rather than poor and angry, roaring approval of a man who manages to connect with facts only about 25% of the time.