Anyone who knows me will chuckle at the accusation that I’m questioning BLS’s labor share data because I’m a closet conservative. The point of this blog – other than self-serving therapy – is to add a small shove in the direction of truth. But maybe I ought to take it as a compliment, coming as it is from a solidly left blog.
Scott Hodge’s Selective Analysis of Debt Drivers Since 2001
Note: I cite some of my work for Pew in this post, but as always, per the disclaimer, I’m only speaking for myself here.
Scott Hodge of the Tax Foundation manages the impressive feat of looking at CBO deficit data since 2001 and then writing this with a straight face:
Urban myths are difficult to put to rest, but perhaps these CBO figures can help dispel the myth that the Bush-era tax cuts contributed to Washington’s current fiscal mess. [Emphasis added]
Well, I at least assume it was with a straight face.
Anyway, notice what a strong claim Scott is making there: not only did the 2001/2003 tax cuts not cause current deficits (a defensible but highly semantic claim), but they didn’t even contribute to them. Not one penny!
Also notice that the chart accompanying the post contradicts Scott’s assertion:
When you add up the legislative tax cut and spending increase factors, you get 28% and 44%, respectively. So yes, spending increases were greater than tax cuts over the last decade! But that still means we’ve had $4 of tax cuts for every $6 of spending increases, relative to CBO’s January 2001 projections. And that ignores some esoteric but highly important caveats, like the fact that CBO classifies refundable tax credits as spending hikes rather than tax cuts.
And what happens if you break down the spending category into component policies? You get Pew’s chart of the same data (full disclosure: I authored Pew’s analysis and designed the chart):
Personally, I’ve never seen convincing evidence that the 2001/2003 tax cuts were responsible for the majority of the shift from projected savings to actual debt that occurred between 2001 and 2011, and indeed, CBO data show that no single policy bears responsibility for more than 13% of this shift.
But that said, the single policy responsible for the greatest share of this shift was the 2001/2003 tax cuts.
(Wonkish aside: Scott and Pew have slightly different percentages because they use slightly different but valid measures of the 10-year shift)
So if Scott is arguing that our debt problem has its roots in many different policies, of which the 2001/2003 tax cuts are just one, then more power to him.
But that’s not what he wrote. Instead, he wrote, “[T]hese CBO figures can help dispel the myth that the Bush-era tax cuts contributed to Washington’s current fiscal mess.” In fact, the CBO data proves exactly the opposite: that among a diverse array of deficit-financed policies ennacted since January 2001 – under both parties and encompassing both spending hikes and tax cuts – CBO data shows that the 2001/2003 tax cuts were the single largest debt driver of the past decade.
Filed under Economics
Just Say “No” To Publicly-Financed Stadiums, Especially in LA
Via the excellent City Ledes, I see that Philip Anschutz is trying to woo an NFL team to Los Angeles (including the Chargers, to which I say, well, don’t bring a knife to that fight, Phil). Of course, LA has been without a team since both the Rams and the Raiders fled in the early 1990s. One positive consequence is that LA is one of the best TV markets in the country for watching American football, what with no blackouts and no risk of local inferiority crowding out the airwaves. Given that NFL ticket prices are now flirting with Disneyland levels and that an NFL stadium only brings eight home games a year to a region (plus the rare windfall from a Super Bowl), I don’t think the average Angeleno family has much to gain from a new stadium. Now, if Anschutz is ready to go all-private in his endeavor, more power to him. But the tendency towards publicly-financed stadiums is unfortunately alive and well, despite notable exceptions such as San Francisco’s AT&T Park, and it looks like Anschutz’s competition has public money on the table:
Leiweke, in confirming for the first time Anschutz’s interest in purchasing a majority share of a team to return the NFL to Southern California after a nearly 16-year absence, Leiweke also said he has spoken with officials from five NFL franchises: Minnesota, San Diego, Oakland, St. Louis and Jacksonville.
Leiweke said he last spoke with an NFL team “a week ago.” That team is believed to be Minnesota. The Vikings are pursuing a $1.05 billion stadium proposal in which Ramsey County and the state of Minnesota would pick up $650 million of the tab.
Minnesota is one of the economically-healthier states, with a 6.5% unemployment rate and no projected FY2012 state government shortfall according to NASBO, but even so I can imagine many more effective uses of $650 million. And it goes without saying that no state or local entity in California would have any business pledging this kind of cash for a private project.
Filed under Economics
Is Labor’s Share of Income at Historic Lows? It’s Not Clear Cut.
Both David Frum and Talking Points Memo are drawing attention to data from the Bureau of Labor Statistics showing that labor’s share of income is at its lowest point since 1947:
The implication here is that workers are capturing a declining share of of our economic output in the form of compensation.
Back in 2004 (which, as you can see from the graph, was a then-historic low according to BLS’s methodology), a paper from the Cleveland Fed critiqued taking these BLS numbers at face value:
Our analysis yields three basic conclusions. First, the “historic lows” in labor’s share are observed only in the nonfarm business sector series produced by the Bureau of Labor Statistics. Other measures of labor’s share—for example, for the nonfinancial corporate business sector or the macroeconomy more broadly—are currently near their averages over the last several decades.
Second, for labor’s share as computed by the Bureau of Labor Statistics, a fall in labor’s share does not necessarily imply a rise in capital’s share; indirect taxes and subsidies constitute a wedge between these two series. Consequently, a fall in labor’s share could be associated with a rise in capital’s share, but it could also be due to a rise in the share of indirect taxes less subsidies. However, we find that the share of indirect taxes less subsidies does not vary much. Further, the terms “capital’s share” and “profit share” are often used interchangeably, ignoring the fact that capital income derives from more sources than just (corporate) profits.
Finally, there is a cyclical pattern to labor’s share: It rises during recessions and falls during expansions. The recent fall in labor’s share (in the nonfinancial corporate sector, and in our broad macroeconomic measure)—back to its historic mean—is typical of the early part of a business cycle expansion. Whether these movements in labor’s share have implications for monetary policy is an open question.
Their analysis is not easily replicable with new data, so it could be that even under the authors’ methodology, labor’s share of income is still at historic lows. But more research and, in the meantime, some skepticism is called for.
American Government is (Probably) More Productive After the Great Recession
Economists observe that productivity often rises during or immediately after a recession. The most logical cause for this phenomenon is that private firms are rational, and prefer to lay off their least productive workers first. I wondered if the same was true of government.
The direct productivity data from the Bureau of Labor Statistics only covers the non-farm business sector. There are a wide variety of reasons for this, among them the fact that it’s difficult to accurately value the non-market goods and services produced by government in a way that’s comparable to goods and services produced by the private sector. In fact, this same (major) caveat applies to the charts I’m about to show you, but they at least provide a first approximation of how government productivity has changed over time.
First, let’s look at real value-added (each sector’s contribution to GDP) per hour worked:
What’s striking (keeping the above caveat in mind) is how much higher government productivity was in absolute terms before 1980, but the private sector surpassed the stagnating public sector in the 1980s.
There are several possible explanations. One is that government has bucked the trend of the rest of the economy and actually gotten worse at what it does over the last 60 years. Now, I have a DMV appointment coming up in two weeks, so believe me when I say I’m primed to be sympathetic to this argument. And in fact I could accept a slower growth rate over time when compared to the private sector, or even a shrinking one insofar as the BEA isn’t capturing people’s actual valuation of public services. But I’m not convinced the public sector has capital “T’” Truly gotten less efficient over time. Not in the age of computers and call centers.
There’s also a more progressive explanation. Much of governments’ “value-added” under BEA definitions is simply employee compensation. That’s because, again, governments “produce” public goods that often cannot be priced in any meaningful way. So the closest proxy of value we can use is what the public sector pays its employees, and so the “public stagnation” you see here may simply reflect stagnating and lower real wages (on average). By the way, valuing output at employment-costs makes sense in a market world where theory predicts that someone’s wage will equal their marginal product, or value-added. But why should we think that governments are any better at valuing its “output” than the BEA (a government agency, I might add) is?
All that said, here’s the same data shown as three-year-annualized growth rates:
What’s interesting here is that even using our flawed measures of government value-added, public sector productivity seems to roughly follow the behavior of private sector productivity following a downturn, especially after 1973 and 2007. So perhaps there are more similarities between the behaviors of public agencies and private firms than we give credit for.
Filed under Economics
Could Repealing the Bush Tax Cuts Pay for Social Security?
TLDR version: Fully repealing the Bush tax cuts would probably more than cover the Social Security shortfall over 75 years. Partially repealing them probably wouldn’t. You would still get debt reduction benefits from either solution, though they would represent a departure from the way we’ve funded Social Security in the past. Meanwhile, present value analysis of government fiscal policy is valid but tricky.
The story begins, as it usually does, with a government report.
Kathy Ruffing of the Center for Budget and Policy Priorities posted an analysis last week of the latest report from the Social Security Trustees. Among her illustrations was a comparison of the present (discounted) value of the Social Security shortfall over the next 75 years to the present value of partially- and fully-extending the 2001/2003 tax cuts over 75 years:
In a subsequent post on the CBPP blog, Paul Van de Water reiterated that CBPP wasn’t proposing funding Social Security with the savings from canceling tax cuts, but rather “illustrat[ing] the hypocrisy of Members of Congress who argue that the tax cuts are affordable but Social Security is not, even though their cost is about the same.”
Megan McArdle of The Atlantic takes issue with the graph, arguing that a) present value analysis is inapplicable to government spending, because “the government does not borrow and save like normal people… The closest it can come to saving is to pay off debt;” and b) 75-year estimates of the costs of extending the 2001/2003 tax cuts are extremely sensitive to your assumptions in the out years.
There’s a lot to unpack here, so let’s start with the salient policy question: by running the scenarios through a Social Security model based on the Trustee’s latest intermediate projections, I found that you could almost certainly pay for the Social Security shortfall by fully repealing the 2001/2003 tax cuts and putting the proceeds into the Social Security trust fund (again, though, this is not what CBPP is proposing).
Estimates of the nominal savings from fully repealing the tax cuts go out until 2021 in the latest CBO Budget and Economic Outlook. The analytical question is how savings grow after 2021. Turns out it’s a moot point when you’re talking full repeal: even if you assume that the savings shrink nominally after 2021 (highly unlikely given rising incomes and inequality), there would be more than enough to keep the system solvent for 75 years (nominal savings from full repeal grow at an average annual rate of 5.9% between 2016 and 2021 according to JCT).
While this solution is a departure from the way we’ve funded Social Security in the past, it’s not nearly as crazy as you might initially think. Assuming that we swap the new cash in the trust fund with Treasury securities, then we still get most of the fiscal benefits of higher revenues: the unified federal budget deficit and publicly-held federal debt decrease from where they otherwise would have been under a current policy baseline (and no, it’s not double-counting). The drawback is that this approach does nothing to restrain Social Security spending.
Also, only partially repealing the tax cuts (that is, only for filers making more than $200,000/$250,000 single/jointly) is probably insufficient to keep Social Security solvent over 75 years. Again, it all depends on your growth assumption after 2021, but it’s telling that while the nominal savings from partial repeal grow at an average annual rate of about 2.5% between 2016 and 2021, it would take a growth rate of 4.6% to keep Social Security solvent through 2085. Remember though that thanks to the inanity of trust fund accounting, this solution would still lead to some reduction in the unified deficit and debt against a current policy baseline.
There’s a political economy consideration with this solution too, which is that Social Security has historically been funded by a revenue source – payroll taxes – that have a clear nexus to future benefits. Backfilling the Social Security trust fund with income tax revenue, which would include revenue on income sources not used to calculated Social Security benefits, would fly in the face of the idea of Social Security being a self-sustaining social insurance program. Our fiscal situation is so dire that such radical ideas ought not to be kept off the table outright, but it should give us pause.
On the more esoteric (but important!) question of using present value analysis in fiscal policy, a few points:
1.) Using present values for fiscal analysis is not the sole volition of CBPP. In fact, CBPP’s 75-year present value estimate of the Social Security shortfall comes directly from the Trustees’ report itself (page 187):
The Congressional Budget Office also uses present values when summarizing their own long-term projections. Here, for example, is a Social Security table from their latest long-term report from last June:
2.) Megan’s just wrong when she says that the “the closest [the government] can come to saving is to pay off debt.” In fact, the Social Security trust fund has saved quite a bit over the years (the draw down of these assets is why Social Security isn’t projected to be insolvent until 2036, rather than 2023 when it starts running annual deficits) and a meaningful part of its revenue between now and exhaustion is projected to come from interest on those savings:
Remember that interest rates are just a measure of opportunity costs, be it the opportunity cost of not saving (hoarding cash) or the opportunity cost of not paying down debt (and thus being saddled with higher interest payments in the future). So Megan’s argument doesn’t really speak at all to the appropriateness of present value analysis in the context of debt-reduction.
3.) That said, the “right” discount rate to use is a tricky question, a classic tricky question for us wonks in fact. The interest rate on debt will undoubtedly rise in the future, even assuming no risk premium or crowding out. In addition, CBO points out that interest on Social Security assets tends to be higher than on publicly-held debt writ large. I have often found that options that are equivalent on a PV basis are not equivalent if you map them out year-by-year. So if the point of the exercise is to look at debt effects, I think it’s more reliable to build a model that dynamically calculates interest payments, and then just run a goal seek scenario to get to your fiscal goal. Of course, that’s not always easy or possible, but when it is possible I find the results far easier to digest.
Filed under Economics
Historical Tax Revenues Are Irrelevant and Insufficient
Roberton Williams has a terrific post at TaxVox that’s worth reading in full. The gist of his argument is that historical revenues as a percent of GDP are, contra Paul Ryan, not a normative benchmark for the government going forward.
It is true, as Ryan’s plan noted, that that level of revenues matches the historical average for recent years. Over the past half century, the share of GDP claimed by federal taxes has ebbed and flowed, ranging from a high of 20.6 percent in 2000 to a low of 14.9 percent each of the last two years. The 50-year average was just a hair under 18 percent (see graph). But nothing says this should be the correct level of taxation going forward. [Emphasis added]
Absolutely right. The population is getting older. We’re no longer fighting the Cold War. The nature of the American economy has shifted dramatically in the last twenty years, to say nothing of the last fifty.
I’d also add this: If you look over the last fifty years – and there’s nothing that makes 50 years the “right” scope (why not 25? Why not 200?) – the weighted average of revenues-to-GDP is 17.8%. The corresponding spending-to-GDP figure over that period is 20.7%. In words, the historical average over the past 50 years has been a deficit of about 3% of GDP. There’s a lot of math involved, but the bottom line you would find is that the nation’s overall debt-to-GDP ratio stays roughly constant when you run deficits of 3% of GDP (remember that the economy grows too, so you can run small deficits and still maintain or even decrease the debt-to-GDP ratio). But if we want to decrease the debt burden, we’ll need to get to deficits below 3% of GDP. Simply reverting to average historical revenues and spending would thus be insufficient to reduce America’s debt burden. And of course, historical averages tell you nothing objective about the cause of the problem. Conservatives like to tout the historical comparison as evidence that we have a “spending” problem, but you could make just as convincing a case that the problem is insufficient revenues (up to today, that is; going forward, there’s far more consensus that the problem is one of runaway health care spending). Really, these statistics are just an inkblot test for one’s prior assumptions.
Filed under Economics
![ernie [dot] tedeschi [at] gmail [dot] com](http://imgur.com/b3LX7.gif "ernie [dot] tedeschi [at] gmail [dot] com")
