Tag economy

The Magically Disappearing Deficit

Danny Vinik 4 Comments

The Congressional Budget Office (CBO) released its 2013 Long Term Budget Outlook today and there’s a lot of good news. Total public debt is projected to hit 100% of GDP in 2038, thanks to growth in entitlement spending and interest payments. However, this number is well below CBO’s estimate last year that public debt would hit 200% of GDP in 2037.*

This is thanks to slightly higher taxes and significantly reduced spending on entitlements and interest payments.

The fiscal cliff deal at the end of last year (officially known as the American Taxpayer Relief Act) made the Bush tax cuts permanent for most Americans and fixed the Alternative Minimum Tax (AMT) to limit its reach. However, the deal also allowed taxes to rise on the wealthiest Americans. Due to that, the CBO now projects that revenues will equal 19.7% of GDP in 2038, up from 18.5% in last year’s report.

On the spending side, two major developments drastically reduced the CBO’s projected spending totals.

First, health care cost growth has slowed considerably over the past couple of years and there is more and more evidence demonstrating that this slowdown is not a short-term result of the recession, but is a permanent bending of the cost curve. This led the CBO to lower its projected health care costs:

A particular challenge currently is estimating the extent to which the recent slowdown in growth can be attributed to temporary factors like the recession or instead reflects more enduring developments. Studies have generally concluded that a portion of the observed reduction in growth cannot be linked directly to the weak economy, and CBO’s own analysis has found no link between the recession and slower growth in spending for Medicare. Accordingly, over the past few years, CBO has substantially reduced its projections of spending on Medicare and Medicaid during the coming decade and slightly lowered its estimate of the underlying rate of growth for health care spending per person for the country as a whole. CBO’s estimate of that underlying rate takes into account spending trends since 1985 but gives greater
weight to the recent experience; because of the pressures to constrain spending growth, the underlying rate is projected to decline gradually in the long run.

The CBO’s 2012 Report projected Medicare and Medicaid spending (plus CHIP and the exchange subsidies) to hit a combined 10.4% of GDP in 2037. In this year’s report, the Budget Office expected those programs to be just 8.2% of GDP. That’s a significant drop.

Second, the extended baseline scenario assumes that sequestration is not repealed, compared to last year’s extended alternative baseline scenario that assumed otherwise. This projection made sense in 2012 when it was widely assumed that Congress would find a way to replace the sequester. But now, sequestration is already in effect and the parties aren’t any closer to finding a replacement. It’s more and more likely that sequester could be here to say. This reduces the CBO’s spending projections significantly:

The Congressional Budget Office (CBO) projects that if current laws generally continued without change, other federal noninterest spending would drop from a total of 11.3 percent of gross domestic product (GDP) in 2012 to 7.6 percent in 2023 and then to 7.1 percent in 2038.

Under the extended alternative baseline scenario in 2012, the CBO projected that spending to be 9.6% of GDP in 2037.

The icing on the cake is that all of this reduced spending will lead to significantly lower debt payments, compared with the CBO’s 2012 projections. Debt payments will still rise from today’s low level of 1.3% of GDP to nearly 5 percent of GDP in 2038 (that’s why it’s a sin we aren’t taking advantage of today’s low rates). But that is much less than the CBO’s 2012 projection of 9.5%.

Having gone through all of that, here’s the overall change in U.S. revenues and spending between last year’s Long Term Budget Outlook and this year’s report:

2013 Long Term Budget

The deficit has dropped by almost two-thirds in the last year alone!

Now, the sequester is still dumb policy and the current projections still leave us with an unsustainable budget (economists and budget wonks agree that we need to get our budget down to around 3% of GDP). But the overall picture is abundantly clear: we’ve already done a huge amount of deficit reduction.

*Note: I’m using the extended alternative baseline scenario from the 2012 Report because it more accurately represents the future policy of both taxes and spending. In this year’s report, I’m using the extended baseline scenario as the Fiscal Cliff deal cleared up the unrealistic assumptions that the CBO used under this scenario in 2012.

The Age Old Question of Technology and Job Loss

Danny Vinik Leave a comment

The Information Technology and Innovation Foundation hosted a debate this morning on whether technological advancement has led to persistent job losses in the economy. The Foundation’s President, Robert Atkinson, faced off against Andrew McAfee, the author of a recent book on the topic and a researcher at MIT’s Sloan School of Management.

McAfee began the debate by running through a number of economic trends that could be linked to technological change. He explained that over the past 30 years, only college graduates have seen growth in wages, while everyone has seen their wages stagnate. At the same time, corporate profits have skyrocketed while labor’s share of income is at all time lows. The top 0.1%’s share of income has increased substantially over the past half-century as well.

These trends, McAfee argued, demonstrate the decoupling of productivity from wage and job growth. In particularly the disruptive advancements in information and technology, most notably with the advent of the personal computer, is a major driver of this decoupling. As productivity has continued to grow, wages have stagnated and the rate of job growth has decreased.

Atkinson pushed back on a number of these points.

“There is zero relationship logically between job growth and productivity,” he said. McAfee presented a correlation between those economic indicators, but had not proven causation. Atkinson joked that it was as if saying that job growth was strong during the 1990s when the New York Yankees consistently won the World Series, but fell off afterwards when the Yankees were no longer winning. But no one would argue that the Yankees recent struggles are the cause of poor job growth. It’s just a coincidence.

Atkinson noted that many people believe technological growth is to blame for poor job growth, because it is an appealing explanation. But these theories only look at first-order effects.

“The reason why we see productivity leading to more jobs is second-order effects,” he said. These second-order effects include such things as new industries that pop up from technological growth or the lower prices that consumers see around the economy, leading to more consumption and stronger job growth in other sectors.

This debate comes up every time job growth falls off, Atkinson said, with each generation thinking that its technological advancement is a new phenomenon wreaking havoc on the labor market.

“We always think our own era of technological progress is the best,” he said. “In fact, it is not the best.”

This was a major point of contention between the participants. McAfee argued that we are rapidly heading for a world where robots replace a large portion of the jobs in the American economy, even for such hard-to-automate positions as airline pilots or doctors. Atkinson could not imagine such a world where airline pilots didn’t exist. They will always have to be on airplanes in case something goes wrong, he said.

But even more than this, Atkinson argued, is that looking at specific industries ignores the effect on technology on the entire labor market.

“You can’t look at individual sectors and extrapolate to the whole economy,” he said.

McAfee emphasized that his analysis was not focused on individual industries, but on the labor market as a whole. Atkinson may be unable to imagine a world with robot airplane pilots, but McAfee would not rule anything out.

“The only thing I’ve learned about technological progress after studying it for a while is never say never,” he said.

Do We Have a Structural Unemployment Problem?

Danny Vinik 1 Comment

That was the question that economists Peter Diamond, Dean Baker and Kevin Hassett debated yesterday afternoon in a panel discussion at the American Enterprise Institute (AEI). Moderated by AEI’s Michael Strain, the panel did not disagree on much, particularly its emphasis on the need for government programs to help the long-term unemployed.

Nobel Prize winning economist Peter Diamond, now professor emeritus at MIT, kicked off the debate by arguing against the oft-repeated claim that the current unemployment problem is not just cyclical, but is structural as well. He focused on the Beveridge Curve, which graphs the unemployment rate compared to the job vacancy rate. It’s shown below:Beveridge Curve

When the unemployment rate is high, the vacancy is rate is low as that generally coincides with recessions when employers aren’t hiring. As you can see from the graph above, the concern amongst economists is that there are now more job vacancies for higher levels of unemployment in the past few years compared with the recent decade. However, Diamond was dismissive of this, nothing that over the long-term, the Beveridge Curve fluctuates dramatically and often crosses back above itself after recessions.

“This is not a tight, technical relationship,” he said. “This is a curve that moves all over the place, in part for reasons we could identify, in part not.

“The path back being above it has happened a number of times before and sometimes after that you stay above,” he added. “Sometimes after that when you get back toward full employment, you’re back to the old curve or even below it. So the issue of thinking about how to interpret the path we’re seeing is something that really calls for digging underneath these aggregates.”

Diamond emphasized that a couple other economic points indicate that this is a cyclical unemployment problem. In particular, the lack of wage growth in any major industry is very surprising if the structural unemployment theory is true. If there was a structural unemployment problem, firms would be unable to find workers with the adequate skills and would have to increase wages to fight for the scarce talent. But that hasn’t been the case, Diamond said. Wages have been stagnant.

He also examined the construction industry in particular to see if the change in long-term unemployed construction workers’ employment has been any different than changes for long-term unemployed workers in other industries. When he looked at the data, he found few differences. This rebukes the idea that long-term unemployed construction workers have been unable to reenter the labor force due to a mismatch in skills.

Dean Baker, the Co-Director of the Center for Economic and Policy Research, added an additional layer to Diamond’s argument, noting that the Beveridge Curve has shifted upwards only for the long-term unemployed, not for the short-term unemployed. This is evidence that there was not a structural employment problem for the long-term unemployed when they were part of the short-term unemployed. The issue began when they became part of the long-term unemployed.

The final panelist, AEI’s Kevin Hassett, focused almost entirely on the problems of those workers.

“When you create a stock of folks who have been unemployed for a long time, then it makes it uniquely difficult to reattach them to the labor market,” he said. “There’s been insufficient attention to the emergency of the long-term unemployed.”

Hassett joked that he’d received a surprising amount of praise from liberal organizations recently for his promotion of government jobs programs to help those workers. Yet, even Hassett in conjunction with liberal economists has been unable to convince policymakers to implement such a program. This, he noted, is devastating to those workers, who see significant negative effects on health and wages due to their long-term unemployment. For those reasons, this is a problem that Congress cannot kick down the road.

The longer we wait to confront this pressing issue, the worse it will become. Unfortunately, the lack of interest from Congress may mean it will get much worse.