:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/56761091/unloading_boxes.0.jpg)
As someone who writes frequently about universal basic income — the idea of giving everyone enough money to live on, no strings attached — the most common argument I hear against the proposal has nothing to do with its cost, or the potential that it’ll discourage people from working, an attack that former Vice President Joe Biden used this week.
The most common criticism I hear, rather, is that basic income would cause massive inflation.
This idea has some intuitive plausibility to it. Imagine the government printed and handed out $1 billion for each person in the country. It seems obvious that this wouldn’t actually make everyone billionaires — that is, it wouldn’t let everyone live in Four Seasons suites and fly Gulfstream jets everywhere. A lot of people would just stop working because of their newfound riches, corporations would have to jack up wages dramatically to keep their labor force, those higher wages would lead to higher prices, and the resulting inflation would wipe out most or all of the gains.
Handing out $12,000 to every American adult every year, as basic income advocates like former Service Employees International Union president Andy Stern have suggested, is a much smaller change than that. But it seems like the same principle should apply. If everyone going to a given Trader Joe’s suddenly has $1,000 more per month to spend, shouldn’t Trader Joe’s jack up prices in response?
Luckily, new research on a program in Mexico gives us a real-world test case for this idea. And it strongly suggests that giving out cash doesn’t cause inflation — or if it does, the effects are very, very mild.
In 2003, the Mexican government launched a food assistance program known as Programa de Apoyo Alimentario (loosely translated to Food Aid Program, or PAL), aimed at poor rural villages. During its rollout, the government did a randomized experiment involving 200 villages. Some villages got in-kind food aid: a bundle of common food items, like rice, vegetable oil, and canned fish. The total market value of the food was about 200 pesos ($20 US) a month per household. Other villages got a similar amount of cash, with no strings attached. Still others got nothing, and served as a control group.
The economists looking at the program — the Naval Postgraduate School's Jesse Cunha, the Geneva School of Economics and Management’s Giacomo De Giorgi, and Northwestern's Seema Jayachandran — wanted to know what the different kinds of transfers did to prices. Did giving out cash, with no strings attached, drive up prices? What did handing out food do to food markets in those villages?
They find no statistically significant change in prices in villages getting the cash transfer. Their estimate is that prices rose 0.2 percent (a tiny change), but they can't rule out the possibility of no effect at all. By contrast, villages getting the food saw prices fall by 3.9 percent. This makes sense: People are less inclined to buy something if they're being given some of it for free anyway. So vendors have to cut prices to get more buyers.
What this means for giving cash versus food
:format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/9283463/pal_map.jpg)
The difference in the effects of cash versus food is driven by poorer, more remote villages, which have fewer merchants and stores. This also makes sense. In a big village with lots of competition between stores, vendors can less often afford to cut prices; they’ve already had to cut them as much as possible to keep a leg up on their competition.
In poorer and more isolated villages, though, stores can effectively function like monopolies and charge a higher markup on their products.
In those, the researchers find that the price decline caused by giving people in those villages food instead of cash enhances the size of the benefit by about 28 pesos ($2.80) per month. The food aid didn’t just help people eat — it made it cheaper for them to buy more food with their own money. In all, researchers estimated, giving food in villages where prices declined was equivalent to a 14 percent increase in the monthly value of the program.
That’s a really intriguing finding. There’s tons of evidence that cash transfers work as well as or better than in-kind transfers, including past research by Cunha on this very program. In a 2014 article on the PAL program, Cunha found that giving people cash led them to buy different kinds of food than the food baskets would've provided, including more nutritious fresh, perishable food that couldn’t be distributed the way PAL food was. He also cites an estimate that the cost of distributing PAL's food aid amounted to 20 percent of the foods' cost, whereas the distribution costs of Mexico's main cash transfer program, Progresa-Oportunidades, amount to 2.4 percent of the cash distributed. Cash was a better match for people’s needs, and it was cheaper to distribute.
Except this paper suggests an important benefit to in-kind transfers, at least in some villages: They push down prices for food and other essentials, while cash transfers don’t. And if that effect outweighs the increased distribution costs of food compared to cash, then giving food is more cost-effective, delivering more aid per dollar, than giving cash.
When I asked the researchers about this, they emphasized that while in some villages that might be the case, that doesn’t necessarily mean that policymakers should encourage giving food rather than cash.
“My takeaway from our paper is less that in-kind is the way to go for remote, uncompetitive villages (especially with improvements in electronic benefits transfers where cash is much cheaper and less prone to corruption) and more a good reminder that poor places are fraught with market imperfections that can limit the effectiveness of cash transfers and we need complementary policies too,” Jayachandran says. “That might be better roads, encouraging store entry through subsidies, etc.”
The Mexican government took that general view too. It has since adapted PAL to be mostly cash but offering in-kind or subsidized fortified milk, which has important health benefits and was underpurchased with the cash program. “In the end, policymakers have to weigh a lot of different factors when deciding which transfer mechanism is optimal (delivery cost, price effects, impacts on consumption, etc.),” Cunha says.
So what does this mean for basic income? Extrapolating from the Mexican case to other countries, particularly richer ones like the US, is tricky. But this research is perhaps the best evidence we have on the inflationary effects of cash programs. And it’s pretty sanguine: While there are interesting differences between giving out cash and giving out food, what definitely doesn’t happen is a cash-provoked inflationary spiral. The research suggests that basic income and related policy really do leave individuals better off, rather than just making goods more expensive.
