The NYT’s Big Deficit Numbers – OpEd

For decades I have harangued reporters about writing down really big numbers, most often budget numbers, without providing any context that would make them meaningful for their audience. Even though leading news outlets, like the New York Times (NYT), Washington Post (WaPo), and National Public Radio (NPR), have well-educated audiences, most readers/listeners have no idea what $250 billion, or some other huge sum, over the next decade means. (Often the time period over which money will be spent is not even specified in a piece – it does matter.)

Few reporters have ever tried to tell me that their audience actually did know the meaning of the very large sums of money that are often discussed in budget stories, when no context is provided. This was explicitly acknowledged some years back in a column by Margaret Sullivan, who was the NYT’s Public Editor at the time. The column includes comments by David Leonhardt, then the NYT’s Washington Bureau Chief, who completely accepted the point.

The piece indicated a commitment to putting numbers in a context that would make them understandable to readers. There’s not much evidence of any follow-up on that one. It is still standard to see budget articles that report the millions, billions, or trillions, with no context whatsoever. It is a safe bet that for most readers, this is the same thing, as David Leonhardt put it, as if they just wrote “really big number.”

The latest item that caught my attention in this respect is a NYT article on the bipartisan infrastructure bill.  The article made a really big deal out of the fact that the Congressional Budget Office (CBO) had scored the bill as adding $256 billion to the debt over the agency’s 10-year budget horizon.

From its treatment in the piece, readers were obviously supposed to believe that this $256 billion is a really big deal. But is it?

If we want some basis of comparison, we can look at CBO’s projection for GDP over this 10-year period. CBO projects that GDP will be a bit over $290 trillion, which means that the addition to the debt it projects will be equal to a bit less than 0.09 percent of GDP over this period.

Alternatively, we can say that 0.09 percent of GDP will be the size of the boost to the annual deficit. If we want a per person figure, the total boost to the debt is projected at a bit less than $800 a head.

We can also express the increment to the debt as a share of current projected spending. That can be found either by going to the CBO budget projections or using CEPR’s “It’s the Budget Stupid” Federal Budget Calculator. That tells us that the increase to the deficit is equal to 0.42 percent of projected spending.

People may disagree on which comparison provides the best context. That is fine, we can experiment with different metrics. Perhaps after some time there will be agreement on what should be the standard, but any of these are clearly much better than just writing down $256 billion.

It is more than a bit mind-boggling that our leading news outlets would insist on a practice that everyone knows is not meaningful to the vast majority of their audience. After all, the purpose of reporting is supposed to be to provide information. Using a really big number with no context, is not providing information. It’s just as if they wrote an article in an obscure language that almost no one in the country spoke. That is not a way to provide information.

Why Confusion on Budget Numbers Matters

We should want budget numbers to be expressed in a way that is meaningful as an end in itself; we want the public to be informed. But getting a clearer understanding also matters for how people view various programs.

Polls have consistently shown that the public hugely overestimates the amount of money going to a wide range of programs, such as food stamps, TANF (Temporary Assistance for Needy Families – the reformed welfare program), and foreign aid. It’s hard to imagine that the public’s support for these programs is not affected by its perception of the amount of money going to them.

If people believe that food stamps take up 20 percent of the budget, they are likely to think very differently about the program than if they realized that it costs roughly 1.4 percent of the budget ($63 billion this year). If the food stamp program actually cost 20 percent of the budget, then people might reasonably think that they could pay lower taxes if we cut it down to size. They might also reasonably think that the program is not very effective, if we could spend this much money and still have serious issues of malnutrition and hunger. The same story would apply to a wide range of other programs.

When I have raised this point with other progressives they almost invariably tell me that people want to believe that these programs cost hugely more than is actually the case because they are racist and want to believe that all their tax dollars are going to undeserving Blacks and Hispanics.

While there are many people who fit this story, exaggerations on the cost of these programs go well beyond the Trump base. There are many people who consider themselves centrists, or even liberals, who also hugely over-estimate the amount of money spent on social programs. Many of these people vote for Democrats and progressive candidates.

It would take a highly-paid DC political strategist to try to claim that these sorts of misconceptions did not affect people’s attitudes towards these programs. In fact, you would have something seriously wrong with your thought processes if you had the same attitude to a program if you believed it was spending $1 trillion a year instead of $63 billion.

Yet, almost none of the liberal/progressive policy groups or funders has ever thought to take on the misconceptions spread by the media. Having worked at and co-directed one of the poorer policy shops in this category, I know that time and resources are scarce, but it is a bit incredible to me that these groups could claim that all the papers, conferences, or workshops they have done over the last three decades have been more important than trying to change the way the media covers budget numbers.

I realize this is asking them to do something new. After all, they all have been writing forever on why the food stamp or TANF budget should be protected or expanded, trying to influence media coverage means doing something different. And, for many of these people doing something different is scary.

Margaret Sullivan wrote her piece as public editor, acknowledging the NYT’s failure to write big numbers in a way that is meaningful to its readers, in response to a petition/e-mail campaign organized by CEPR, Fairness and Accuracy in Reporting, Media Matters, and Just Foreign Policy. You may notice some big-name liberal Washington policy shops missing from the list.

Needless to say, we got zero funding. Changing budget reporting and how people see the world just is not on the agenda of big liberal funders. Anyhow, I don’t think it is impossible to change the way the media talk about the budget. Everyone knows their current reporting is incredibly irresponsible. It just takes some pressure to force the issue.

Why Worry About Government Deficit/Debt?

The media largely take it as a given that we should view government deficits and debt as a bad thing. I imagine most of the reporters writing this stuff would struggle to come up with an answer if anyone asked them “why?”

There is a classic Econ 101 story that we can tell about the evils of budget deficits. The story runs that when the government borrows money, it puts upward pressure on interest rates. Higher interest rates then crowd out investment (and net exports – slightly longer story). And less investment means less productivity growth, which means that we will be poorer in the future because of our deficits today.

The big problem with this story is that interest rates have been extraordinarily low during the last year and a half as the deficit has exploded. The interest rate on 10-year Treasury bonds has been under 2.0 percent for the whole period, and in recent weeks it has been under 1.5 percent. By contrast, in the glory days of budget surpluses in the late 1990s, the 10-year Treasury rate was over 4.0 percent and sometimes over 5.0 percent. The story of deficits leading to high interest rates doesn’t appear very credible just now.

There is a story about budget deficits causing inflation. That clearly can be an issue, and there is some evidence inflation could be a problem now. But, at this point it is difficult to sort out the effects of disruptions associated with an economy reopening after a pandemic from the effects of an overheating economy. Most of the uptick in inflation that we have seen thus far has been due to rising new and used car prices, which in turn are largely the result of a semi-conductor shortage due to a fire at a major plant in Japan.

But most of the complaints get to the burden of the debt on our children. The idea is that we are passing on this massive debt, which will be a crushing burden on our children as they attempt to live their lives and raise their own kids. This story usually comes with emphasizing the size of the debt in trillions, which is of course a very big number, but one that has almost no meaning for anyone.

If we’re being serious about the burden of the debt, the first thing to note is that we don’t have to pay off the debt. We just have to pay the annual interest on the debt. This debt service is the actual burden of the debt.

By this measure we don’t have much to worry about at the moment. Our net interest on the debt is currently around $230 billion a year, or roughly 1.0 percent of GDP. (This takes out the $80 billion that the Federal Reserve Board refunds to the Treasury each year from the interest on the bonds it holds.)[1] By comparison, the debt service burden was over 3.0 percent of GDP in the early and mid-1990s.

The deficit hawks usually respond that this story could change if interest rates were to rise to more typical historical levels. That is true, although it’s not clear that a sharp rise in interest rates is very likely. Furthermore, even if we did see interest rates rise to something like 4-5 percent, it’s far from clear this is any sort of disaster story. Remember, the 1990s was a very prosperous decade, in spite of relatively high debt service burden.

Also, we have to remember that the bulk of interest payments are made to other people of the same generation. If we think of some distant future, all of us who are building up the debt today will be dead. The people who hold the bonds and collect interest will be the children and grandchildren of people alive today. If the debt service is placing a burden on the budget, we can just increase the taxes on these lucky people who are collecting the interest payments. That is not a burden across generations, this is a question of intra-generational equity.[2]

If Payments on Government Debt Are Bad, How About Rents on Government-Granted Patent Monopolies?

I realize that I am just about the only economist who ever makes this point, but I am more interested in being right than agreeing with other economists. Direct payments are only one way the government pays for things, it can also pay for things by granting patent and copyright monopolies.

The deal with these monopolies is that the government tells individuals or companies to innovate or do creative work, and then we will give you a monopoly. The government will arrest anyone who competes with you, allowing you to charge a far higher price than in a free market.

The difference between the patent (or copyright) monopoly price and the free market price is the rent that the company is able to charge as a result of this government-granted monopoly. The gap is often quite large. In the case of prescription drugs, the patent protected price may be more than a hundred times the free market price. Drugs are almost invariably cheap to manufacture and distribute, it is patent monopolies that make them expensive.[3]

The amount of money that we pay out each year as a result of patent and copyright rents is enormous. I calculated that it is over $400 billion annually in the case of prescription drugs alone, almost twice the debt service burden. Adding in medical equipment, computer software, and other areas where these rents can be a large share of the price, the total can easily come to more than $1 trillion annually, more than $3,000 for every person in the country.

It makes zero sense that we would worry about the burden created by government debt, and the resulting debt service, but pay zero attention to the burdens created by government-granted patent and copyright monopolies. As noted above, direct spending and the granting of monopolies are alternative modes of payment by the government. The deficit hawks only want us to look at the costs from the first one.

In some cases, the trade-offs are made explicitly. The Food and Drug Administration wanted more drug companies to perform pediatric clinical trials to ensure that their drugs were safe and effective for children. Rather than having the government pay for these trials directly, drug companies can extend the length of their patents by six months if they conduct a pediatric trial.

If the government just paid the companies to conduct the trials, there would be an item in the budget that would add to the deficit and debt that we are then supposed to be concerned about. But when the government just says that we will give you a longer patent monopoly, the deficit hawks say this is fine – no cost.

That sort of thinking may make sense at the New York Times, Washington Post and other high-end news outlets, but it makes zero sense in reality land. I opt to continue to live in the latter.

The Debt and the Planet

As I’m sitting here in Southern Utah, we are facing a multi-year drought, hundred degree temperatures, severe water shortages (also our source of power), and are inundated with thick haze from the forest fires in California. It’s a bit hard for me to see the debt as the major injustice facing future generations. We are destroying so much of the nature that makes this country and the world beautiful or even livable.

I keep envisioning the following scenario for 2050: one of today’s leading deficit hawks, now up in years, boasting to a group of young people about how we paid off the national debt. The world outside is scorched, with few trees or any other plant life. Most of the animals that are alive today have gone extinct. Coral reefs are ancient history.

Somehow, I can’t imagine the kids being grateful. This again is not a complex concept. We will hand down a whole natural and social world to our children and grandchildren. The burden of the debt and the debt service is such a trivial part of this picture, it’s hard to believe serious people would waste their time on it.

Notes:

[1] Arguably, we should be looking at the real interest burden of the debt, which would subtract out the extent to which the real value of the debt is reduced by inflation. If we applied this standard, the real debt service burden would be negative, since the inflation rate current exceeds the interest rate.  (The real interest rate is the nominal interest rate minus the inflation rate.)

[2] There is an issue of foreign holders of government debt. The interest on this debt is a burden on the country, but that goes well beyond government debt. Insofar as foreigners hold any U.S. financial asset – stock, real estate, the bonds of private corporations – the payments create a burden for the country. In the late 1990s, even as we had large budget surpluses, foreigners were buying up large amounts of U.S. financial assets. There is no direct relationship between the size of our government debt or deficit and the amount of U.S. financial assets being purchased by foreigners.

[3] Patent monopolies also create perverse incentives. The high mark-ups created by these monopolies give companies incentive to mislead doctors and the public about the safety and effectiveness of their drugs in order to maximize sales, as happened with the opioid crisis.

This first appeared on Dean Baker’s Beat the Press blog.