By Dean Baker
Polls consistently show that the public hugely overestimates the share of the budget that goes to items like SNAP (food stamps), Temporary Assistance to Needy Families (TANF) and foreign aid. People will typically give answers in the range of 20 to 30 percent of the budget for these categories of spending. In reality, the shares are 1.5 percent for SNAP, 0.4 percent for TANF and 0.4 percent for foreign aid.
I would argue that this matters, since the public’s willingness to support a program depends in part on how much they think we are spending on it. This is for two reasons; the first is simply that people are only willing to pay a limited amount in taxes to help the poor here and abroad. If they already think they are spending a lot for this purpose, they will be reluctant to spend more.
The other reason is that people will reasonably be concerned about the efficiency of the programs. If all our tax dollars are going to help poor people, and yet we still have so many people in poverty, then our anti-poverty programs must not be very efficient. If that is the case, added additional dollars probably will not do much to help the poor. Nor will modest cuts do much to harm them.
All of this seems pretty straightforward and not really debatable, yet when it comes to educating the public on the true size of these programs, interest is very close to zero. That is hard to understand, especially when the route to a better-educated public is pretty easy to see.
The most obvious reason that people grossly overestimate the amount of spending on these programs is that their budgets are always discussed as billions of dollars. No one knows how much billions of dollars are, except that it means lots of money.
Discussing budget numbers in millions, billions and trillions is incredibly irresponsible reporting. It is the job of the media to be informing their audience. Writing that food stamps cost $70 billion a year, or that TANF costs $20 billion, is not informing readers. It is just putting down numbers, equivalent to a mindless fraternity ritual, that serves no informational purpose.
This is infuriating, because it is not hard for reporters to put these numbers in a context that would be meaningful to most of their audience. If they made a point of expressing these numbers as a share of the budget, people would immediately know whether these items are a big deal or small change in terms of the whole federal budget.
While I know that many reporters work hard and are not anxious to get an extra task assigned to them, this one is pretty simple. It just means dividing whatever number they are looking at by the total budget. There is no reason it should take more than a few seconds.
I have been haranguing reporters and editors about this one for a long time. None has ever tried to contest the basic point, and argue that more than a tiny minority of their audience has any idea how much $70 billion a year is in terms of the federal budget.
Of course, sometimes it gets even worse. Budget numbers are often expressed over multiple years, with the reporters not even being clear whether the figure being discussed was a one-year or ten-year number. I have encountered educated, intelligent people who thought the $1.5 trillion price tag on the 2017 Republican tax cut was a one-year figure. In fact, it is a ten-year figure that comes to a bit more than 0.6 percent of projected GDP, and 2.7 percent of the federal budget.
The most explicit acknowledgment of this problem came in a column from Margaret Sullivan, who was then the public editor of the New York Times. She took up this issue in response to a public campaign by Just Foreign Policy, Media Matters and FAIR. She strongly agreed that the paper often printed numbers that were meaningless to most readers. Her piece included a quote from David Leonhardt, who was then the Washington editor for the paper:
“The human mind isn’t equipped” to deal with very large numbers. When people see these numbers, he said, they read it as “a lot of money” or “a really big number.”
Sullivan and Leonhardt both agreed that the paper needed to put numbers, especially big budget numbers, in a context that would be understandable to readers. Incredibly, in spite of the strongly worded comments of the paper’s public editor and Washington editor, nothing changed.
I have tried, with virtually no success, to try to enlist liberals and progressives in an effort to pressure media outlets to do what both Sullivan and Leonhardt agree they must do if they want to serve their audience: put numbers in context. I am not sure why there is such an intense lack of interest. There are two obvious possibilities.
- People may believe that there is little hope of success. The media for some reason simply will not express budget numbers in a way that make them meaningful to their audience.
- It wouldn’t make any difference in political outcomes. In other words, the fact that the public hugely overestimates the share of the budget that goes to various social programs makes no difference in determining their political viability.
Taking these in turn, it is a bit difficult to imagine that reporters have some principled objection to writing budget numbers in a way that is meaningful to the people who read or hear them. I find it hard to believe that most reporters actively want to confuse their audience. I also find it hard to believe that there is some dark conspiracy among editors, producers, publishers and other management personnel and ownership to keep the public in the dark on budget matters.
There is an issue that it is more work for reporters, but we are not talking about major time commitments here. It is really hard to believe that reporters could have that big an objection to take the few seconds needed to do some simple arithmetic on a spreadsheet.
There really is no other side on this one. There simply is no good reason to not put numbers, and especially big budget numbers, in a context that is meaningful to readers. There is no serious dispute that almost no one can make any sense of these large numbers without any context.
This brings up the second reason not to try to push media outlets on this issue—that it wouldn’t make any difference anyhow. The argument here is something to the effect that we have lots of people who hate these programs because they are racist. They want to believe that all their tax dollars are going to undeserving black and brown people, and they are not going to let the facts get in their way.
This is surely in part true. As has become painfully obvious in recent years, there is a substantial segment of the electorate that does not want to be bothered with facts. We can assume that this share of the population is somewhere between 25 and 35 percent. For these people, hitting them over the head with the fact that only a small portion of the budget goes to anti-poverty programs is not going to make any difference in their attitude towards these programs.
There are also people that believe that the government has the responsibility to ensure everyone has a decent standard of living. These people would be willing to support anti-poverty programs even if they did impose a substantial tax burden. Let’s say this group is also 25 to 35 percent of the population.
Then there is a middle group that is willing to support anti-poverty programs to a limited extent. They are not willing to pay any price to keep people out of poverty, but if these programs are effective, they would be willing to pay a somewhat higher tax bill.
For this group, the fact programs like TANF and SNAP are only a small share of the budget is likely to make a considerable difference. If we give this group credit for some amount of serious thought, if they believed, as many do, that anti-poverty programs account for 40–50 percent of the budget, they can hardly be blamed for not wanting to see them expanded or objecting to cuts. After all, if we’re spending half the budget trying to pull people out of poverty, and we still have so many poor people, then these programs must not be very effective. How could we think that spending 5 or 10 percent more or less would make any noticeable difference?
I realize that people are not constantly searching for information on the size of the TANF or SNAP program, but if they were constantly hit over the head with the fact that these programs are relatively small shares of the budget, it is likely to sink in. This knowledge might also benefit friends and relatives who argue with these people because they think we should help the poor, even though they wrongly believe that these programs are a large share of the budget.
In fact, being continually hit over the head with the fact that TANF is 0.4 percent of the budget may even affect the attitudes of some people who hate these programs. Some of them may still have the view that they dislike even a small share of the budget going to people they see as undeserving, but think it is more important to support politicians who will ensure that they have healthcare and their kids can afford college.
If the battle to change media reporting on budget numbers is both winnable and one that is likely to have a considerable impact on national politics, why is there so little interest among liberals and progressives in pursuing this fight? I don’t have a great answer to this, but I would say at least 90 percent of the story is inertia.
People are used to yelling about evil Republicans and pro-corporate Democrats who do the bidding of the rich. They are also used to targeting corporations for bad practices, such as ripping off workers on pay and benefits or polluting the environment. There is not much tradition of harassing the media on bad reporting practices.
Asking liberal and progressive groups to take this on as a cause is asking them to do something they have not previously done. Unfortunately, for many on the left, this is hard.
There is a joke in Washington that the motto among liberals is, “We’ve been losing for 40 years, why change now?” That motto explains a lot.
A version of this post appeared on CEPR’s blog Beat the Press (3/28/19).