Friday, December 14, 2012

Lie of the Year? "$1.5 trillion spending cuts already passed"

For my undergraduate degree I majored in philosophy, and one of the ideas I picked up then that has stuck with me is that the "coherence" theory of truth has a lot to recommend it.  It answers many of the various objections that epistemological skeptics make and that subjectivists make to objectivists while retaining the primacy of logic and rejecting the metaphysical premises behind moral relativism.  To oversimplify, you might not agree with someone else's world view, but you still have to keep your own world view coherent, and once it is established that both you and another person both believe in certain basic principles, if you disagree on some derivative matter one of the two of you is more wrong than the other.

The moral status of abortion may be an example of an issue whereby the level at which disagreement begins is too fundamental for either side to prove the either side to be definitively wrong.  There just aren't many commonly held premises, even when drilling down to the metaphysical level.  One ends up appealing to a "correspondent" theory of truth, i.e. arguing that the other side's beliefs do not correspond to reality.  If one person's view of reality happens to be entirely materialist (that nothing, not even consciousness, exists apart from matter and energy) while the other is metaphysical dualist, the disagreement will be pretty fundamental.

But most debates between "subjectivists" and "objectivists" are over matters of degree within a coherence construct.  If the coherence test is applied to such an extent that the "edge" of its applicability circumscribes all of human consciousness, and that metaphysical premise itself forms a common cornerstone, then notions of "subjectivity" drop out as irrelevant, à la Wittgenstein's beetle in a box.  There might be another universe out there, but we are living in this one.

I write this as a preamble to making three comments about the U.S. economic and political debates.  Allow me to take as the first example what PolitiFact called the "Lie of the Year": the Romney campaign's Jeeps and China ad.
 How effective a criticism of PolitiFact's decision is it to note that every word of the ad is actually true?  Presumably not very, since that's held to not be the issue.  At the time, Politifact said, "In this fact check, we examine whether the sale of Chrysler came at the cost of American jobs" instead of examining what the ad actually claimed.  According to PolitiFact, the ad "presents the manufacture of Jeeps in China as a threat, rather than an opportunity to sell cars made in China to Chinese consumers."  Now that IS true, it is presented as a threat on the basis of the well established economic principle of opportunity cost (corporate resources spent on expanding plant in China are resources not spent on expanding plant in the U.S.) but the question is why outsourcing suddenly became an "opportunity" when Politifact treaded lightly on the Obama campaign's many charges that Romney was running to become Outsourcer in Chief.  Where is the coherency?  Now it's true that Romney could be challenged on his own "coherency" in choosing this line of attack by noting his track record at Bain.  But that's actually a charge of hypocrisy, not a truth claim.  As Ryan Chittum at the Columbia Journalism Review put it:
[The ad is] saying that Chrysler’s Italian owners “are going to build Jeeps in China.” But happens to be true, even if it was happening before 2009 under its German and private-equity ownership. Cars made overseas by an American company (even one with Italian owners) are cars that won’t be made in the U.S., and it’s fair to say those jobs are outsourced.
On the other hand, it’s high hypocrisy for Romney the free-trader private-equity guy to attack anyone for outsourcing production...

Glenn Kessler of the Washington Post dodged the fact that what the ad said was true by declaring that "The series of statements in the ad individually may be technically correct, but the overall message of the ad is clearly misleading."  Kessler assigned his most damning designation of Four Pinnochios on this basis, yet he assigned zero to Obama's claim in his second debate with Romney about his communications regarding Benghazi which were only technically correct and technically was not even technically correct.  The Columbia Journalism Review was founded by Victor Navasky, who has used his publication The Nation to, amongst other things, argue that Alger Hiss wasn't a spy for Stalin despite the overwhelming evidence to the contrary.  The CJR's Chittum nonetheless points out a mitigating factor re the Jeeps and China ad, namely that it started with a Bloomberg story with problematic wording and then was amplified inaccurately by a Washington Examiner blogger (before actually being dialed back in the Romney ad transcript from what the Examiner said).  No such mitigating factor exists with respect to Obama's misleading claim in the second debate.  Instead of getting "caught up in the liberal echo chamber," Obama's decision to create the impression that he called Benghazi a terrorist attack from Day 1 was a decision to create something, not picking up to forward on a liberal attack line already in circulation. Politfact, of course, also took a pass when it came to rating Obama's remarks, instead seeing the back and forth with Romney at the second debate as an opportunity to mark down Romney, calling his criticism of Obama on the point "half true."  The bottom line here is that it's a mug's game to try and expose "fact checking" bias by pleading the "facts."  One has to make an appeal to coherency, not correspondence.

There are other examples one could go into.  Romney's infamous "47%... pay no income tax" remark was taken to task by fact checkers in part because "nearly two-thirds of households that paid no income tax did pay payroll taxes."  Yet when it comes to trimming the growth of Social Security benefits, were these benefits paid for by tax revenue?  In that case, the payroll taxes are spun as being insurance policy payments or otherwise "earned" benefits; in other words, when the question is whether the 47% are carrying their share of the load, their payroll taxes are deemed to be building up the public pot, but when the question is limiting S.S. payouts, these same contributions are deemed to be building up a private entitlement.

Now the example I meant to talk about today (but which I've taken a while getting to) goes back to my last post about the extent to which U.S. unemployment is cyclical or structural.  In the context of the "fiscal cliff" debate the New York Times today claims that "Obama already agreed to more than $1.5 trillion in cuts last year."  This $1.5 trillion number comes courtesy of the Center on Budget and Policy Priorities, however it should be noted that these "cuts" nonetheless still not only allow "discretionary" programs to continue to grow with inflation, they allow for a further $65 billion in spending above and beyond that over the next decade.  How do you spin spending that exceeds inflation (and more) as "cuts"?  By reaching back to inflated 2010 appropriation levels and using that as the baseline.  The 2010 appropriation bills were actually adopted in 2009 when the demand on the social safety net was near its height and when Democrats enjoyed a filibuster-proof majority in the Senate in addition to their majority in the House, except for some special bills in 2010 for disaster assistance, border security, and the Patent and Trademark Office which the CBPP of course also included in order to further inflate its baseline.  The CBPP says that the 2010 appropriation "simply reflects the level that Congress deemed appropriate" at the time.  Well of course.  You could say the same thing about 2007, or 2013 (we're already more than than two months into the 2013 fiscal year), but of course then the claim of $1.5 trillion in "cuts" would collapse.

Now what would be an appropriate baseline?  The answer to that is what is most coherent with the view you have taken elsewhere.  Dean Baker, for example, has repeatedly insisted that the "graph that gives a better picture of the problem of the budget deficit in relation to the economy" is one that calls attention to projected deficits in January 2008 (I've copied his graph here).  The blogger Kevin Drum insists on using a graph that he cuts off at 2008 in order to argue that "Washington Doesn't Have a Spending Problem." Drum says he cuts off his chart because "numbers in the chart have spiked over the past four years because the recession has temporarily depressed GDP and temporarily increased spending, but that spike will disappear naturally as the economy recovers".  Yet Drum elsewhere tallies up "Discretionary spending cuts already passed in 2011: $1.5 trillion"  No "natural disappearance" here, it's "cuts already passed" by Congress!  Not passing another disaster relief bill like in 2010 becomes passing a cut.  Coherency here would mean tallying up the change in spending since 2008, although that would of course completely undermine the point about all the "cuts" that have occurred.

My last two blogposts about the U.S. prior to this one were about whether the current U.S. economic situation is the "new normal" and whether Obama's remarks at the second presidential debate were misleading or not.  People can disagree with the former and say that 2007 should be the touchstone, but if so, don't elsewhere start calling a level of federal government spending that was decided in 2009 the appropriate reference point.  People can disagree about my negative take on Obama's claim in October about what he said on September 12 by saying his words were narrowly true, but don't elsewhere say that Romney was a liar because while his ad was narrowly true it's what viewers were lead to believe that matters.  

Tuesday, December 4, 2012

the new normal: U.S. economy already nearing potential

The main argument that's been trotted against the "deficit scolds" is that Washington should wait until the U.S. economy has recovered.  When will it have recovered?  When it is approaching "full employment."

In economics, "full employment" is a rough synonym for "potential GDP" and it refers to the underlying productive capacity of the economy.  Potential GDP is the blue line in the graph below.  Although the U.S. emerged from the officially determined recession some time ago (indicated by grey shading), according to Keynesians a  large "recessionary gap" or shortfall in the aggregate demand for goods and services in the economy remains, as indicated by the gap between potential GDP and actual GDP (which is given in red):

Although there's no doubt about the red line, just where the blue line should be plotted is less than an exact science.  In this case, the blue line is plotted by the Congressional Budget Office or CBO, and what is of particular note here is that the CBO has been repeatedly shifting the line towards the right (i.e. towards the red line), as Brad DeLong has graphed, to the effect of steadily reducing the demand deficiency.  DeLong notes that the CBO has drawn down its potential GDP track by 8.5% since 2007, well in excess of the 2008-2009 contraction, which Dean Baker has pegged at 4%.  Here's a graph that has the potential GDP line coming very close to actual GDP:

This particular graph, which the President of the St Louis Federal Reserve, James Bullard, has been using, has come under a lot of attack by Keynesian economists because it challenges the hypothesis that there is currently a significant output gap.  The red dashed line here is generated by using a "Hodrick–Prescott filter," which is named after economists Robert J. Hodrick and Edward C. Prescott, who developed it to separate the "cyclical" element from the "trend" element.  Now an HP filter has various limitations, such as working with less information near its end point, but Mark Thoma takes issue with what it says in the middle: "the HP filter reveals a period of substantial above trend growth through the middle of 2008. This should be a red flag for Bullard. If he wants to argue that steady inflation now implies that growth is close to potential, he needs to explain why inflation wasn't skyrocketing in 2005. Or 2006. Or 2007."  One could start here by pointing to the fact that Paul Krugman has defended the use of data that uses a Hodrick–Prescott filter by arguing that a gap with potential GDP need not be defined exclusively by inflation.  Does Thoma believe that because the general consumer price index in Japan rose less than 4% between 1985 and 1989, the Japanese economy was not operating above potential despite real GDP soaring more than 20% during those four years?  Isn't there an obvious analogy between Japan's equity and commercial real estate bubble bursting in 1990 and the U.S. residential real estate bubble bursting in 2008?  Growth cannot always be below trend, meaning that unless Thoma is going to define a decade as the short term, he needs to acknowledge occasions where it has been above trend.

Notwithstanding his 2009 use of the IMF's HP filtered data, in July Krugman weighed in on the debate over Bullard's graph to note that back in 1998 he took issue with those who used a HP filter to conclude that Japan's economy was close to potential that year.  What Krugman doesn't tell you is that he took issue wrongly, claiming back then that "in retrospect it will seem clear that Japan's 1998 output gap was 8 percent or more... so that demand-side policies to close that gap are of very real importance."  In fact the OECD, which marked down Japan's potential growth to 1.6% in 1994 and forecast then that the estimated output gap for Japan in 1997 was, in Krugman's view, "remarkably small:  -  1.2  percent," has been vindicated.  According to the IMF's model (that would be the very same model Krugman used in 2009 to support his claim about the output gap in the U.S.) Japan's output gap for 1998 was less than two percent.  While Krugman argued that Japan was facing a prolonged slump in GDP growth below potential, the IMF correctly interpreted Japan’s lost decade as a slowdown in potential.  I might add here that the IMF also says that the U.S. had an "inflationary gap" (that is, a negative output gap) over 1% in 2005, 2006, and 2007.  Despite having wrongly prescribed aggressive "demand-side policies" for Japan in 1998, Krugman is currently pounding the drum for them to be applied in the U.S. in 2013.  

Ironically, Krugman exhibited a moment of worry in February 2008, noting that a HP filtered graph of productivity was flashing warning signs.  Productivity is a key component of potential GDP, meaning that he saw back then an explanatory factor for the U.S. economy's low growth that wasn't coming from the demand side.  Here is a graph of the year-over-year change in potential GDP as measured by the CBO:
Note the slump from 2000 to 2011.  The CBO optimistically believes the trend will reverse this decade, but if it doesn't, the CBO will continue to shrink its projected output gap by ratcheting down its potential GDP projections.

There are a variety of other metrics that one can point to that are inconsistent with the thesis that there is a large output gap, including the fact that inflation is running at 2%, pretty much right on the average since the late 1990s, the employment cost index (wages and benefits) is rising, and  U.S. capacity utilization has recovered to pre-crisis levels. 

But a Canadian now working for the St Louis Fed, David Andolfatto, makes a particularly compelling argument for deeming current U.S. unemployment structural as opposed to cyclical:
The Canadian unemployment rate is is blue, the U.S. rate in red.  Throughout the 80s and 90s, Canada's unemployment rate was consistently higher.  Although the Canadian unemployment rate rose around the 2001/2002 recession and again in 2008/2009, the rise was smaller than in the U.S., and Canada reformed its unemployment insurance program in the 90s, restricting benefits.  If it took Canada three decades to move in front of the U.S., why should we believe the U.S. will quickly reduce its unemployment to Canadian levels?  It's more likely that most of the "business cycle" adjustment has already occurred, such that remaining adjustment is subject to long term processes.

This brings us to yet another string of arguments, namely those advanced by Casey Mulligan.  Mulligan goes beyond the fact that the share of 25 to 54-year-old non-college educated men in the work force has trended down for decades to note how U.S. policy has encouraged non-participation in the labour force in recent years by easing eligibility rules for unemployment insurance, increasing the generosity of food stamps, etc:
The American Recovery and Reinvestment Act, popularly known as the stimulus, gave unemployment insurance recipients a weekly bonus, and offered to pay for the majority of their health insurance expenses. FDIC and Treasury reduced some “unaffordable” mortgage payments, which means that successful people need not apply. The list goes on and on.

The essential consequence for all of these is the same: a reduction in the reward to activities and efforts that raise incomes.

Dean Baker and Paul Krugman have occasionally taken issue with Mulligan, although Krugman frequently suggests that he considers Mulligan too much his inferior to engage.  At one point Krugman says "Mulligan and others keep emphasizing examples of individual groups that have managed to gain jobs by cutting wages or offering other attractions to would-be employers."  Why is this inconsistent with Keynes?  Because Keynes' was of the view that if you increased the skill level of all the unemployed in the economy in one swoop or reduced the wage cost to the employer of hiring them, this wouldn't reduce unemployment when there be an output gap because the problem is not with the supply of labour but on the demand side.  Krugman's reply is that a sub-group that reduces its unemployment says nothing about whether the whole group ("all unemployed in the economy") could realize the same success.  Everybody can't be an above-average potential hire, observes the Krugman.  I don't find this at all convincing, however.  Krugman ought to be be chastising Dean Baker instead of citing him if he were consistent.  Why?  Because Dean Baker's #1 policy prescription after the economy has returned to potential is a lower U.S. dollar.  I should think it would be obvious that not every country can depreciate its currency at the same time.  At issue here is the international competitiveness of the United States; if a particular group in the U.S. can take actions that improve its employability the Keynesians need to explain why it wouldn't work for the entire country to take a similar action.  

To be sure, I think Mulligan goes too far when he seems to suggest on occasion that the cause of the recession is to be found on the supply side.  But that's not the question at issue here on the doorstep of 2013.

UPDATE December 6:
Job vacancies continue to rise: