I have posted an Excel spreadsheet which is an extension of one included in a zip file posted by Herndon, Ash and Pollin (hereafter called HAP). I will use data from that spreadsheet to look at the HAP criticisms of the RR paper.

A good starting point is the now infamous Excel spreadsheet shown to the right.

After noting the coding error by which five rows were excluded, the first question that occurred to me was “where’s the beef?”. The number of countries on which the 90 percent threshold is based is a mere seven (Belgium having been left out by RR). The HAP critique points out that RR is using only 71 data points (110 after Belgium and 14 other excluded data points are added). Since there were relatively few data points, my first inclination was to try to look at that data so see how it was distributed. The following table shows the 71 data points used by RR, the 25 data points for Belgium, and the other 14 excluded data points (marked by ^):

Real GDP Growth for Debt of ≥90 Percent of GDP for 1946–2009

* Columns excluded by RR because of Excel error
^ Data excluded by RR

Note that the 1951 data point for New Zealand is the only data point used by RR for that country. The data points from 1946 to 1949 for New Zealand are excluded. Tab A of the aforementioned spreadsheet shows the following numbers for New Zealand from 1946 to 1955:

Debt/GDP and Real GDP Growth for New Zealand: 1946–1955

^ data excluded by RR

As can be seen, New Zealand had real GDP growth of 14.7 percent the year before and 4.3 percent the year after the –7.6 percent growth in 1951. Hence, had RR used a debt threshold of 85 percent of GDP, the average rate of growth would have been 3.8 percent. On the other hand, if RR had used a debt threshold of 95 percent of GDP, New Zealand would have had no data points. Both of these calculations assume that RR continues to exclude 1946 through 1949. If these are included, the averages become 4.6 and 5.1 percent, respectively. Only by using a threshold of 90 percent of GDP and excluding 1946 through 1949 did RR come up with –7.6 percent. A cursory inspection of the data, as I have done here, would have shown the –7.6 figure to be an unrepresentative outlier, requiring some sort of corrective action.That cursory inspection also turns up the real GDP growth of –10.9 percent for the U.S. in 1946. This was the year after World War II ended so it’s no surprise that real GDP dropped that year. If that year had been excluded, the average real GDP growth for years that the U.S. was above the 90 percent threshold would have been +1.0 percent instead of –2.0 percent.

One might argue that Belgium’s real GDP growth of 15.2 percent in 1947 was also an outlier. Excluding that year, however, merely drops the average real GDP growth for years that Belgium was above the 90 percent threshold from 2.6 percent to 2.0 percent. This is because the effect of an outlier is much greater in a small country sample when the weighting is done per country as was done by RR.

The following table shows the effect on the average real GDP growth for all years above the 90 percent threshold of various corrections and changes in calculation method:

Average Real GDP Growth for Debt ≥90 Percent of GDP for Various Scenarios

* Column excluded by RR due to Excel error

^ Data excluded by RR

The first line shows the slightly negative figure calculated by RR (Reinhart and Rogoff). The next 4 results are in response to various corrections suggested by HAP. The second line shows an improvement of 0.3% when the infamous Excel error is fixed. The biggest improvement of 1.3% occurs in the third line when the missing years of 1946 to 1949 are included for New Zealand. Another improvement of 0.3% occurs when the missing years of 1946 to 1950 are included for Australia and Canada. Finally, another improvement of 0.3% occurs if the data is weighted by country-year instead of by country as suggested by HAP. This gives a real GDP growth of 2.2 percent given by HAP in the Abstract on page 1 of their critique.The final 5 lines show that, even if one insists on weighing the data by country instead of country-year, any reasonable attempt to minimize the effect of outliers will bring the result much closer to HAP’s value of 2.2% than RR’s value of –0.02%. The first of these shows that simply excluding the outlier of New Zealand will cause an improvement of 1.2 percent. The second shows that another improvement of 0.5% occurs if one excludes the outlier of the U.S. in 1946.

A more consistent way of dealing with the small samples of debt right after World War II would be to exclude all countries with no case after 1951. This would raise the real GDP growth to 2.0 percent, nearly as high as the HAP value of 2.2 percent. The same result would occur if the starting year of the data was moved from 1946 up to 1952. The final line shows that, if the study were to simply focus on the “modern era” after 1980, the result would be a nearly-as-high 1.8 percent. As can be seen, any of the fixes beyond merely fixing the Excel error would bring the calculation much closer to HAP’s value of 2.2 percent than RR’s value of –0.02 percent.

What is the lesson to be learned from all of this?

As mentioned here and here, the Reinhart and Rogoff paper was not peer-reviewed. However, the latter article makes the following interesting comment:

So the answer is to only accept peer-reviewed work as economic knowledge, right? Nope. That would be a) too limiting, and b) wouldn’t advance the epistemological cause as much as you think. Peers have their own sets of biases, particularly as gate keepers.

I do think that peer-review does have a role to play, but I can see that it’s not the only answer. Nonetheless, it does seem that we could at least clearly label what is peer-reviewed and what is not. I don’t recall the absence of peer-review having been mentioned once during the public debate of the past three years. Only once the Excel error was found did we hear, “Oh yeah, that paper was never peer-reviewed”.

There is an additional step that I think would help a great deal. For any paper to be taken seriously, the authors should have to give their sources and show their work. In their online appendix, RR stated:

We took great pains to provide the data in as accessible form as possible, including especially meticulous source documentation in the spreadsheets, far more than one sees normally posted with journal papers. So we are simply stunned when bloggers and irresponsible commentators say we have not shared our debt data. Open access to our data has been central to our whole project.

I believe that RR is referring to the links to data that they have posted here. However, it seemed that HAP had to go through a great deal of effort to recreate RR’s numbers. It was only by requesting the original Excel spreadsheet that they were able to start recreating the numbers. Speaking of the spreadsheet, I was unable to find a copy of the spreadsheet anywhere on the web even now. I manually keyed in the numbers that appeared in the one graphic that I saw on the web and added it as tab C in my spreadsheet. I then tried to reproduce the numbers in the spreadsheet using RR’s sources but was unable to find all of the data. Fortunately, HAP posted the zip file with their spreadsheet and I was able to replicate the numbers with that.

Why don’t economists just post their original spreadsheets? There may be many reasons. Some may be willing to put up with peer-review when required but not want every pointy-headed number-cruncher with too much time on their hands to be going through their work. I suspect that those number-crunchers would catch a number of errors or questionable methodology that peer-review would not catch. In addition, some economists might feel that publishing their spreadsheets would reveal trade secrets and/or help other economists to compete in their area of study. Hence, I think that it’s largely up to those who consume the studies to demand that the work be released. Of course, economists would still be free to release studies without peer-review and showing precisely how they arrived at their conclusions. But if those studies were simply ignored or treated as interesting ideas until they can be verified, I suspect that many of those economists would be willing to “show their work”.

Related articles

• Researchers Finally Replicated Reinhart-Rogoff, and There Are Serious Problems
• GOP’s go-to economics study debunked
• Statistic cited to defend austerity partially based on Excel error
• Europe should thank Silvio Berlusconi
• Meet the 28-year-old Student Who Exposed Two Harvard Professors Whose Shoddy Research Drove Global Austerity

Hello, my name is Marshall and I have some issues with your article “Effect of Reagan, Kennedy, and Bush Tax Cuts on Revenues”. I found some issues here, mainly involving the Reagan Tax cuts. You say

the GDP reached a high ten-year growth rate of 35.2 percent from 1983 to 1993. However, it reached higher highs of 37.5 percent from 1992 to 2002, 45.71 percent from 1947 to 1957, and 50.28 percent from 1958 to 1968. In fact, the … graph shows that the ten-year growth rate in GDP has been relatively stable since 1975 to 1985, though it began to drop in 2008 and is projected to stay weak through 2015. Hence, these figures don’t provide any strong evidence that the Reagan tax cuts permanently affected the GDP one way or the other.

As Marshall states, his comment actually involves my analysis titled “Effect of Reagan, Kennedy, and Bush Tax Cuts on Revenues”. The text he quoted above is from the end of the section titled “Effect of Reagan Tax Cuts on Revenues and GDP — Long Analysis”. To give it more context, here is the full paragraph from which he quoted:

Hence, the evidence is that the Reagan tax cuts decreased revenues over what they would have been, at least over the short (ten-year) term. The only remaining argument in favor of the Reagan tax cuts, at least from a revenue point of view, would seem to be that they permanently raised the level of the GDP, thus bringing in slightly higher revenues far into the future. According to the graph and second table, the GDP reached a high 10-year growth rate of 35.2% from 1983 to 1993. However, it reached higher highs of 37.5 from 1992 to 2002, 45.71% from 1947 to 1957, and 50.28% from 1958 to 1968. In fact, the above graph shows that the 10-year growth rate in the GDP has been relatively stable since 1975 to 1985 though it began to drop in 2008 and is projected to stay weak through 2015. Hence, these figures don’t provide any strong evidence that the Reagan tax cuts permanently affected the GDP one way or the other.

Marshall continues:

Now, there are some issues with this:

Yes, GDP growth was very consistent from the 40s to the 70s (though there was a slight decline in the 70s), and the 80s (January, 1980-December, 1990) from what it seems didn’t change the GDP rating all that much compared to any of these decades. However, it’s worth noting that the GDP growth of the 90s was nowhere in comparison really (January, 1990-December, 2000). When looking at the mean the 80s number seems to be in the ballpark of about 4.6 or 4.7 percent growth in GDP; however for the 90s it seems to be about 3.8 or 3.9, somewhere in between these numbers, so overall looking at that it does seem that Reagan’s tax cuts may not have had any real impact but Clinton’s tax increases (along with Bush 41’s increases) didn’t seem to be all that beneficial to GDP growth. My overall point being that tax cuts may not be that bad especially when compared to tax increases.

here’s the website displaying GDP growth rate. It’s interactive as well.

Can you provide your source for the above GDP growth numbers? They do not agree with the numbers that I have seen.

The graph to the right is from my last article titled “Do Tax Cuts Increase Economic Growth?”

The blue, yellow, and red lines show the one-, five-, and ten-year annualized changes in the quarterly real GDP. The graph shows that the ten-year annualized change was relatively stable from 1974 to 2008, remaining between two and four percent for that entire period. Much of that apparent stability was because ten years just happens to be close to the length of several of the past business cycles. Recall that the three recessions prior to 2008 were in 2001, 1990–91, and 1980–82. That pattern was broken by the deep recession of 2008-09.

The graph does not show GDP growth to have been noticeably better during the 80’s than the 90’s. To get a more accurate measurement, it is better to look at GDP growth over entire business cycles. The following table shows real GDP growth over all business cycles (taken from trough to trough) since 1949 as determined by the National Bureau of Economic Research:

Real GDP Growth by Business Cycle

* 2012, quarter 4 has not been identified as a trough. It is just the most recent quarter.

Some of the business cycles are arguably too short to provide meaningful data. The following table combines the shorter business cycles to remedy this:

Real GDP Growth by Business Cycle (shorter cycles combined)

* 2012, quarter 4 has not been identified as a trough. It is just the most recent quarter.

As you can see in the table above, the two business cycles from 1980 to 1991 had an annualized GDP growth rate of 3.09 percent versus a slightly higher 3.39 percent during the business cycle from 1991 to 2001. The first table shows that the annualized GDP growth rate of the single business cycle from 1982 to 1991 was a slightly higher 3.74 percent. In any case, all of these growth rates are about the same and no where close to the 4.6 or 4.7 percent growth rate that you quote for the 80’s above. Where did those numbers come from?

Marshall continues:

You also asked for examples of tax cuts being beneficial.

Actually, I said the following in my analysis:

There may well be valid arguments in favor of tax cuts. But higher tax revenues does not appear to be one of them.

Then in the article to which he replied, I concluded with the following:

As I mentioned, I am yet to find a single economic study that purports to show evidence that any income tax cut has ever paid for itself. If anyone who reads this should know of one, please leave a comment with a link to that study. Thanks.

In any case, Marshall continues:

I have some links to provide, but I’d like to say two things beforehand:

1. These links are obviously in favor of lower taxes; they are from the Heritage Foundation, an obviously conservative site and also I believe it’s called the Freedom and Prosperity Institute or something like that
2. I’m not sure if you’ve read them or not but they are worth looking in to also one of them is a video series which should be even easier though the video doesn’t give any direct links it does refer to the IRS “SOI” which is something

That being said, here are the links:

 

 

Heritage
Freedom and Prosperity

Both of those articles are written by Daniel J. Marshall. In fact, I wrote an article titled How to Mislead with Statistics, that references the second linked article. In it, I said

I first became familiar with Dr. Marshall writings when he worked for the Heritage Foundation. The first one that I remember seeing was titled The Historical Lessons of Lower Tax Rates. Following is a chart from that article:

The numbers in this chart are similarly not corrected for inflation. The simplistic argument that “tax revenues nearly doubled in the 1980s” (as stated in the title) prompted me to post an analysis titled Effect of Reagan, Kennedy, and Bush Tax Cuts on Revenues. Following is the short analysis at the beginning of the article:

The argument that the near-doubling of revenues during Reagan’s two terms proves the value of tax cuts is an old argument. It’s also extremely flawed. At 99.6 percent, revenues did nearly double during the 80s. However, they had likewise doubled during every single decade since the Great Depression! They went up 502.4 percent during the 40s, 134.5 percent during the 50s, 108.5 percent during the 60s, and 168.2 percent during the 70s. At 96.2 percent, they nearly doubled in the 90s as well. Hence, claiming that the Reagan tax cuts caused the doubling of revenues is like a rooster claiming credit for the dawn.

Furthermore, the receipts from individual income taxes (the only receipts directly affected by the tax cuts) went up a lower 91.3 percent during the 80s. Meanwhile, receipts from Social Insurance, which are directly affected by the FICA tax rate, went up 140.8 percent. This large increase was largely due to the fact that the FICA tax rate went up 25 percent, from 6.13 to 7.65 percent of payroll. The reference to the doubling of revenues under Reagan commonly refers to TOTAL revenues. These include the above-mentioned Social Insurance revenues for which the tax rate went up. It seems highly hypocritical to include these revenues (which were likely bolstered by the tax hike) as proof for the effectiveness of a tax cut.

Hence, what evidence there is suggests there to be a correlation between lower taxes and lower revenues, not higher revenues as suggested by supply-siders. There may well be valid arguments in favor of tax cuts. But higher tax revenues does not appear to be one of them.

This is followed by a much longer analysis that corrects the numbers for inflation and looks at a number of other factors. In any case, the above chart in Dr. Marshall’s paper is followed by the following chart:<

I wrote the following about this chart in 2005:

Chart 11 is titled “Lower Tax Rates Work: Revenues Grew Faster Under Kennedy and Reagan” and shows the average annual increase in real income tax revenues for five time spans. I believe that the time span 1962–1969 was meant to be 1962–1968 as that most closely matches the budget numbers. The table below shows the numbers from the chart and my own calculations from the 1997 U.S. Budget.

First Year	1953	1962	1969	1977	1981	1990
Last Year	1961	1968	1976	1980	1989	1995
Number of years in span	9	7	8	4	9	6
Revenue Growth (chart 11)	0.01	4.79	1.53	—	2.2	1.34
Revenue Growth (budget)	0.11	4.67	1.55	7.48	2.19	1.56

In looking at Chart 11, the first glaring question is what in the world happened to the time span 1977–1980!? Is it missing because it had a 7.48 percent growth rate, very much confusing the issue? In addition, why are the time spans of varying lengths?

Marshall continues:

Conservative site:

This one I myself am a bit skeptical of but he/she does provide a lot of links

These are the videos. They are a three-part set about the Laffer Curve. They also go into detail about the Reagan Tax Cuts (somewhat):

The first video goes into detail mostly about the Laffer Curve theory but talks about Tax Cuts and some fantasies about them.

This one reviews Reagan’s tax cuts more in depth though again not all that much but it’s still worth looking into.

The third video doesn’t go into that much detail about the Tax cuts so i won’t post it, though if the links work then you should be able to find it rather easily in the “other videos” list.

I haven’t had a chance to look closely at the conservative blog that Marshall mentioned. However, I noticed that it includes a similarly simplistic graph as the one above, showing the growth in receipts during the 80s without correcting for inflation or comparing the growth to other periods. I did watch the two videos he linked to and I noticed that both of them are narrated by our good friend, Daniel J. Marshall, after he moved to the Cato Institute. His arguments seem similar to those that he made in the articles he wrote at Heritage. They tend to be simplistic and incomplete (some might say cherry-picked). In any case, I would not depend on him as one’s chief source.

Marshall concludes:

I myself am not very familiar with economics. … From what I have learned, tax cuts do not always pay for themselves. But there are arguable examples of them doing so (e.g., the Reagan tax cuts). Personally, I think tax cuts pay for themselves only when the cut goes from 70 to 28 like it did with Reagan or when it goes from some obscenely high rate to a low rate. Regardless, these are my sources. If I find more I’ll post them.

As I said in my analysis of the Kennedy tax cut,

it would seem possible that Kennedy’s tax cut was beneficial, at least in reducing this oppressive top marginal rate but that Reagan’s tax cuts took the marginal rates to a level where the negative effects far outweighed any positive effects.

Hence, I would say that the tax cut from 91 to 70 percent under Kennedy may have had overall beneficial effects. The data as I can read them are not clear on whether or not this tax cut lost revenue. However, a Washington Post article titled “Where does the Laffer curve bend?” asked a number of economists and most of them estimated the bend point to be between 50 and 70 percent. So if one accepts the Laffer curve in its simplest form, a tax cut to a rate above the bend point would increase revenue but a tax cut below that bend point would lose revenue. Of course, as suggested by the post to which Marshall replied, there are likely other factors involved and the Laffer curve, while possibly a useful concept, is overly simplistic.

Related articles

• Do Tax Cuts Increase Economic Growth?
• Boehner Misleadingly Claims Tax Revenue at a Record High
• Jeffrey Sachs: Professor Krugman and Crude Keynesianism
• Esquire:How Not to Make America Great
• Want to Fix the Deficit? Get Real — Bloomberg
• Crude Sachsism

Related articles

• Us Economy Shrinks: Gdp Falls 0.1% in Q4
• Graphed. Why the budget is bleeding
• Sequestration Will Slow Real GDP Growth — But Not Because of Demand-Side Effects
• GDP Drop

David Kopel in his interview.

As I mentioned last time, David Kopel, a professor at the University of Denver, was on the December 17th edition of The PBS Newshour. Regarding the assault weapons ban, he said the following:

Related articles

• Did the Assault Weapons Ban Work?
• Feinstein has assurance assault weapon ban will get a vote — Reuters
• Joe Biden: “Assault Weapons” Aren’t Used by Criminals to Commit Crimes, So No One Needs Them
• Former Speaker Newt Gingrich dismisses call for assault weapons ban as ‘propaganda’ — @thehill
• Senate overrides Quinn’s assault weapon ban maneuver
• Joe Biden: Shotgun Beats Assault Rifle in Case of Earthquake!

In the third quarter, corporate earnings were $1.75 trillion, up 18.6% from a year ago, according to last week’s gross domestic product report. That took after-tax profits to their greatest percentage of GDP in history.

But the record profits come at the same time that workers’ wages have fallen to their lowest-ever share of GDP.

“That’s how it works,” said Robert Brusca, economist with FAO Research in New York, who said there is a natural tension between profits and the cost of labor. “If one gets bigger, the other gets smaller.”

Many of the stories also included a graph which looked like the following:

The actual numbers and sources for this and the following graph can be found at this link. As can be seen, corporate profits are at their highest percentage of GDP since 1947 when this series began. Likewise wages and salary accruals are at their lowest percentage of GDP.

On a related topic, some other stories have pointed out a gap between the growth in productivity and real hourly compensation. For example, an August 10th Washington Post Wonkblog post stated the following:

The other part of Romney’s claim — that wages and employment track productivity — is actually false. Unpublished data from BLS, generously provided to me by the Economic Policy Institute’s Larry Mishel and Nicholas Finio, shows that wages tracked productivity growth until about 1970. After that, wages stagnated even as productivity continued to grow.

In fact the Bureau of Labor Statistics (BLS) has published data that shows this phenomenon. This data was used to create the following graph:

The red line is labor productivity (output per hour) and the blue line is real hourly compensation, both for nonfarm business. The indices were obtained from the BLS web site as described at the bottom of this sidebar page and adjusted so that the base year (with a value of 100) was 1947 instead of 2005. Note that real compensation closely tracked productivity until the early 1970s. Compensation did stagnate somewhat until the late nineties and, after about a decade of increased growth, has stagnated again. Productivity, however, stagnated only briefly in the mid seventies and early eighties and has grown briskly since then. This has caused an increasing gap between compensation and productivity.

This increasing gap is shown by the purple line which is the ratio of productivity to compensation. This ratio was fairly stable from 1950 to 1970, grew slowly until about 1982, and has grown fairly briskly since then.

In addition, this gap is addressed in a January 2011 essay in the Monthly Labor Review titled “The compensation-productivity gap: a visual essay”. That essay looks at the gap during several periods that contain multiple business cycles. The following table shows the annualized change in productivity, compensation, and their ratio for these periods using the data in the prior graph. It also shows the average annual values wages and corporate profits for these periods using the data in the first graph.

Productivity did ease up a bit in the seventies and eighties but remained above one percent per year even during these periods. Compensation, however, was below one percent per year during the 1970s and 1980s, and since 2000. This has caused the compensation-productivity gap to widen fairly quickly since 1979.

The aforementioned essay goes into more detail on some of the causes of this widening gap:

There are two components that account for the gap between real hourly compensation growth and productivity growth. The first is the difference between the price indexes used to account for inflation in the BLS productivity and hourly compensation measures. The second is the change in “labor share,” which, as explained earlier, is the share of output that is accounted for by workers’ wages, salaries, and benefits.

Before 2000, the difference between the growth rates of the CPI and the IPD—that is, the difference in inflation rates—explained most of the gap in each period. For 2000 to 2009, an unprecedented decline in labor share accounted for most of the gap.

The underlying causes and significance of these two factors are debatable and require further study. What is clear, however, is that wages have been under pressure since the 1970s and this pressure is not explained by a lack of productivity or corporate profits.

Related articles

• Labor’s Lost Leverage
• Corporate profits hit record as wages get squeezed
• Corporate profits have soared while wages plummet — a story in two graphs
• Two Lines Diverge in a Labor Graph
• Corporate profits are highest-ever share of GDP, while wages are lowest-ever
• America Needs a Raise: The Case for a Higher Minimum Wage

Illustration of the 1994 Assault Weapons Ban (from USA Today)

The Sandy Hook Elementary School shooting in Newtown, Connecticut has brought calls to reinstate an assault weapons ban like the one that expired in 2004. From a round table discussion of the shooting on the December 16th edition of This Week With George Stephanopoulos:

WILL: In 1996, a man went into a gym class in Scotland, killed 16 five– and six-year-olds and the teacher. A few years ago in Norway, a young — deranged young man killed, what, 69 people on an island, mostly teenagers. Connecticut has among the toughest gun laws in this country. Didn’t help. Scotland and Norway have very tough gun laws. Didn’t help. So…

STEPHANOPOULOS: Didn’t stop, but it does lessen the occasion of violence, doesn’t it?

WILL: Yeah, why don’t…

EDWARDS: And, George, since Columbine, there have been 181 of these school shootings.

WILL: We did — remember, we did — we did have a ban — we did have a ban on assault weapons. When we put the ban in place, these incidents did not really decline in a measurable way. And when we took it off, they didn’t increase in a measurable way.

George Will is saying that the ban on assault weapons did not have a measurable effect on these mass-shootings. On the December 17th edition of The PBS Newshour, David Kopel, a professor at the University of Denver, was more explicit, saying the following:

Well, I think we can look at what happened when she had her ten-year in the past. The Congress, when it enacted that ban, also ordered that a formal study be done of the results of it.

The study was performed by the Urban Institute, a very well-respected, somewhat left-leaning think tank in Washington, D.C., and the Urban Institute reported that it had no effect on homicide rates. There was no statistically significant benefit in terms of saving lives.

Kopel’s statement that the study had reached such a definitive conclusion sounded a little surprising. Looking online, I found the Urban Institute study titled “Impact Evaluation of the Public Safety and Recreational Firearms Use Protection Act of 1994 — Final Report”. The overview of the study concludes:

At best, the assault weapons ban can have only a limited effect on total gun murders, because the banned weapons and magazines were never involved in more than a modest fraction of all gun murders. Our best estimate is that the ban contributed to a 6.7 percent decrease in total gun murders between 1994 and 1995, beyond what would have been expected in view of ongoing crime, demographic, and economic trends. However, with only one year of post-ban data, we cannot rule out the possibility that this decrease reflects chance year-to-year variation rather than a true effect of the ban. Nor can we rule out effects of other features of the 1994 Crime Act or a host of state and local initiatives that took place simultaneously. Further, any short-run preventive effect observable at this time may ebb in the near future as the stock of grandfathered assault weapons and legal substitute guns leaks to secondary markets, then increase as the stock of large-capacity magazines gradually dwindles.

We were unable to detect any reduction to date in two types of gun murders that are thought to be closely associated with assault weapons, those with multiple victims in a single incident and those producing multiple bullet wounds per victim. We did find a reduction in killings of police officers since mid-1995. However, the available data are partial and preliminary, and the trends may have been influenced by law enforcement agency policies regarding bullet-proof vests.

The following pages explain these findings in more detail, and recommend future research to update and refine our results at this early post-ban stage.

This sounds very different from Kopel’s definitive “no statistically significant benefit in terms of saving lives”. Yes, they were “unable to detect any reduction to date in two types of gun murders that are thought to be closely associated with assault weapons” but they did detect a “6.7 percent decrease in total gun murders between 1994 and 1995″ and a “reduction in killings of police officers since mid-1995″. It is important to note that this study was published on March 13, 1997, just over two years after the ban took effect on September 13, 1994. Hence, the study itself recommends “future research to update and refine our results at this early post-ban stage”.

In fact, two additional studies were published. Like the Urban Institute study, key authors were Christopher S. Koper and Jeffrey A. Roth. The latter of the two, published in June, 2004, is titled “An Updated Assessment of the Federal Assault Weapons Ban: Impacts on Gun Markets and Gun Violence, 1994–2003″. From its first chapter:

The Ban’s Success in Reducing Criminal Use of the Banned Guns and Magazines Has Been Mixed

• Following implementation of the ban, the share of gun crimes involving AWs [assault weapons] declined by 17% to 72% across the localities examined for this study (Baltimore, Miami, Milwaukee, Boston, St. Louis, and Anchorage), based on data covering all or portions of the 1995–2003 post-ban period. This is consistent with patterns found in national data on guns recovered by police and reported to ATF.
• The decline in the use of AWs has been due primarily to a reduction in the use of assault pistols (APs), which are used in crime more commonly than assault rifles (ARs). There has not been a clear decline in the use of ARs, though assessments are complicated by the rarity of crimes with these weapons and by substitution of post-ban rifles that are very similar to the banned AR models.
• However, the decline in AW use was offset throughout at least the late 1990s by steady or rising use of other guns equipped with LCMs [large capacity magazines] in jurisdictions studied (Baltimore, Milwaukee, Louisville, and Anchorage). The failure to reduce LCM use has likely been due to the immense stock of exempted pre-ban magazines, which has been enhanced by recent imports.

It is Premature to Make Definitive Assessments of the Ban’s Impact on Gun Crime

• Because the ban has not yet reduced the use of LCMs in crime, we cannot clearly credit the ban with any of the nation’s recent drop in gun violence. However, the ban’s exemption of millions of pre-ban AWs and LCMs ensured that the effects of the law would occur only gradually. Those effects are still unfolding and may not be fully felt for several years into the future, particularly if foreign, pre-ban LCMs continue to be imported into the U.S. in large numbers.

The study states that gun crimes involving AWs (assault weapons) did decline, primarily due to a reduction in the use of APs (assault pistols). However, this decline was offset by steady or rising use of other guns equipped with LCMs (large capacity magazines). The study concludes that “the ban’s exemption of millions of pre-ban AWs and LCMs ensured that the effects of the law would occur only gradually”. Once again, this sounds very much different from Kopel’s statement that the assault weapon ban provided “no statistically significant benefit in terms of saving lives”.

This points to a common problem with interviews on news shows, even excellent shows like the PBS Newshour. It is very easy to make many types of erroneous statements with little risk of having the errors exposed, at least not in real time. Even though there may be one or more experts with opposing views, none of them will have all of the facts of all studies at their fingertips. It is therefore easy to slip in various erroneous numbers or conclusions with little risk of being contradicted. This suggests that it might be useful to have similar interviews online where each statement could subjected to some minimal fact-checking before the conversation continues. In any case, it shows that such statements should be treated simply as possibilities unless and until they can be verified.

Related articles

• “Feinstein to Introduce Updated Assault Weapons Bill in New Congress” (volokh.com)
• Everything you need to know about the assault weapons ban, in one post (washingtonpost.com)
• Dems Move to Reinflict “Assault Weapons” Ban (moonbattery.com)
• Reviving a gun law that didn’t work (sfgate.com)
• Expert On Gun Regulation Says “Assault Weapons” Bans Are Useless (openmarket.org)

As I mentioned, I have yet to find a single economic study that purports to show evidence that any income tax cut has ever paid for itself. If anyone who reads this should know of one, please leave a comment with a link to that study. Thanks.

I received a couple of replies to that request but neither of them focused on income tax cuts in the United States. One focused on capital gains tax cuts and the other focused on countries with relatively weak tax authorities where tax cuts might increase compliance (like Russia). In any case, I replied to both. I also have not found an economic study that shows an income tax cut paying for itself from any other source. However, I am continuing to search for such a study and have posted links to any related information that I’ve found at this link. I believe that only one of those links is to a study that purports to show a tax cut that paid for itself. It is a paper published by Laffer Associates titled “The Onslaught From The Left, Part III: The Capital Gains Tax”. From the paper:

We now have at least three years of data to assess the effect of the 2003 capital gains tax rate reduction on capital gains tax receipts. Figure 6 shows that from 2002 through 2005, capital gains receipts have doubled, from $49 billion to $97 billion. The latest projection for receipts in 2006 is $110 billion. This represents a 124% increase in revenues over four years with a 25% cut in tax rates. You can’t ask for more than that.

As stated, this paper involves a capital gains tax cut and looks at just three years worth of data. Figure 2 on page 4 of this Congressional Research Service paper shows that capital gains tax receipts sank back toward their 2002 level in the 2009 financial crisis.

In any event, I took those studies that gave actual numbers for the estimated revenue recouped following tax cuts and compiled them into a table at this link. Following is that table:

Estimates of the Percent of Revenue Cost Recouped after Tax Cut

As you can see, none of the studies projected that tax cuts would pay for themselves. At best, they projected that about a third of the cost of labor taxes and half of the cost of capital taxes would be recouped. In fact, the CBO study suggests that the feedback could actually be negative, causing the revenue loss to be larger than the static revenue loss. Following is a summary of the study given by Bruce Bartlett:

A 2005 Congressional Budget Office study during the time that Republican Doug Holtz-Eakin was CBO director concluded that a 10 percent cut in federal income tax rates would recoup at most 28 percent of the static revenue loss over 10 years. And this estimate assumes that taxpayers have unlimited foresight and know that taxes will be raised after 10 years to stabilize the debt/GDP ratio. Without foresight and no compensating tax increases or spending cuts, leading to an increase in the debt, feedback would be negative; i.e., causing the revenue loss to be larger than the static revenue loss.

This seems to agree with the study by Christina and David Romer, mentioned in a prior post on my blog, which found that “deficit-driven tax increases” have had a positive effect on economic growth. In any case, I am interested in any studies that give estimates for the revenue cost recouped following tax cuts. This especially applies to any studies that suggest that an income tax cut in the United States would pay for itself. As mentioned, I have yet to find such a study.

Related articles

• The Fiscal Games (logarchism.com)
• The Bush Era Tax Cuts Didn’t Create The Wealth They Were Supposed To (businessinsider.com)
• Deal Flow on the Rise: Capital Gains Tax to Blame? (axialmarket.com)

In this paper we examine the tradeoffs between rates, tax expenditures, and the progressivity of the tax schedules that are inherent in revenue-neutral tax returns. We show that plans that advance steeply lower marginal tax rate structures would require deep cuts in tax expenditures to offset the revenue losses arising from low rates. Because many of the largest tax expenditures benefit middle– and lower-income households, deep reductions tax expenditures can alter the distribution of the tax burden. To illustrate these tradeoffs, we examine as an example a set of tax rate reductions specified in Governor Romney’s tax plan. We show that given the proposed tax rates and proscription against reducing tax expenditures aimed at saving and investment, cutting tax expenditures will result in a net tax cut for high-income taxpayers and a net tax increase for lower– and/or middle-income taxpayers—even if individual income tax expenditures could be eliminated in a way designed to make the resulting tax system as progressive as possible.

The Romney campaign has disputed this conclusion; both Romney and Ryan have pointed to “six studies” which support the Romney tax plan. A blog posting on the Romney website states that “six independent studies have proven there are sufficient upper-income expenditures to lower individual tax rates, protect the middle class, and make the tax code more pro-growth”. 

There has also been a great deal written outside the Romney campaign on these six studies. Two examples are a PolitiFact article titled “Ryan says six studies say the math works in Romney tax plan” and an article titled “Wonkblog’s comprehensive guide to the debate over Romney’s tax plan”.

Because of the large number of conflicting articles on the Tax Policy Center study and the six studies that countered it, it seems that a first step was to organize those articles in one place. I’ve posted links to the studies and related articles at this web page. It contains the following:

1. The original Tax Policy Center article and some follow-up and earlier articles.
2. The “six independent studies” critiquing the Tax Policy Center article, along with articles critiquing specific studies.
3. Other articles favorable of Romney Tax Plan.
4. Articles critiquing the “six independent studies”.
5. Other articles critical of Romney Tax Plan.
6. IRS Tax Data.

The IRS Tax Data is included because the PolitiFact article states that the Feldstein and Rosen studies used 2009 tax data. More precisely, the initial Feldstein study states that it is “using the most recent IRS data, which is based on tax returns for 2009 and published in the current issue of the IRS quarterly publication” and the Rosen study states that it relies on “summary data from the IRS’s Statistics of Income (SOI) for tax year 2009, the latest year for which such published data are available”. Neither give the precise sources, links to it on the web, or the work by which some of their figures were derived from this data. This makes it difficult to verify the data since there is a great deal of IRS data out there (as can be seen by the sources that I provide). Poorly sourced articles have long been a pet peeve of mine. I can’t help but think that they are poorly sourced so as to make them more difficult to verify and critique.

In any case, I was able to verify most of Feldstein’s numbers. The following table shows those numbers:

Projected Revenue Gain/Loss from Romney Tax Plan
(billions of dollars)

*estimated by Feldstein (else 2009 IRS data)

Sources:

• 2009 IRS Tax Data, Table 1.4, Sources of Income, Adjustments, and Tax Items
• 2009 IRS Tax Data, Table 2.1, Sources of Income, Adjustments, Itemized Deductions
• Martin Feldstein’s Wall Street Journal article, August 28, 2012
• Martin Feldstein’s blog post, September 02, 2012

In Feldstein’s Wall Street Journal article, he states the following:

The key question raised by the Romney plan’s critics is whether this revenue loss can be offset by broadening the tax base of high-income individuals. It is impossible to calculate the exact effects of the future reforms since Gov. Romney hasn’t specified what he would do. But refuting the Tax Policy Center’s assertions doesn’t require that. It only requires knowing if enough revenue could be raised from high-income taxpayers to cover the $186 billion cost.

A little later on, he states:

And what do we get when we apply a 30% marginal tax rate to the $636 billion in itemized deductions? Extra revenue of $191 billion—more than enough to offset the revenue losses from the individual income tax cuts proposed by Gov. Romney.

So Feldstein’s basic argument is that the revenue loss of $186 billion shown in the second to the rightmost column above is more than offset by the $191 billion revenue gain shown in the rightmost column. The first step in testing this contention is to verify Feldstein’s numbers. I could not verify the numbers followed by asterisks in the IRS data. However, those numbers have a relatively small effect on the final numbers.

According the Feldstein, the “$49 billion was from taxing dividends and capital gains at reduced rates that would not be subject to further reductions”. This reduces the estimated revenue loss by $10 billion. The $15 billion is from “eliminating the tax on interest, dividends and capital gains for married couples with incomes below $200,000, and for single taxpayers with incomes below $100,000″ and increases the estimated revenue loss by that amount. Finally, the $33 billion reduction in the estimated revenue loss from $219 billion to $186 billion is due to Feldstein’s assertion that “past experience shows that taxpayers do respond to lower marginal tax rates by acting in ways that increase their taxable incomes”. Regarding this assertion Washington Post columnist Ezra Klein says the following:

Feldstein assumes fairly large, and very positive, growth and behavioral effects from the tax cuts. But he doesn’t assume negative effects. Most models — including, as I understand it, TPC’s — assume that as you cut deductions, taxpayers who were managing their finances to take advantage of those deductions stop doing that. That makes the deductions effectively worth less money, and makes it harder to pay for tax cuts.

In any case, Feldstein’s other numbers closely match the IRS numbers except for one. He gives the “income-tax revenue in 2009 before all tax credits” as $953 billion. The IRS data gives this as $976 billion for all returns and $950 billion for all taxable returns. However, this only has an effect of $4 billion on the projected revenue loss. Even with this, the projected revenue loss of $190 billion is just covered by the projected revenue gain from base-broadening of $191 billion.

There are some issues with this projected revenue gain of $191 billion, however. First of all, Feldstein mentions in his follow-up blog post that critics have pointed out that the “30 percent marginal tax rate is too high for these taxpayers because of the 20 percent Romney rate reduction”. He then states that “using a 25% marginal tax rate instead of 30% would reduce the revenue from eliminating deductions by 5% of $636 billion or $32 billion”. This cuts the projected revenue gain to $159 billion.

Next, Feldstein’s figures are based on the idea of eliminating all deductions for taxpayers whose adjusted gross income (AGI) is $100,000 or more. But when asked about the $100,000 limit in a September 14 interview with George Stephanopoulos, Romney said the following:

No, middle income is $200,000 to $250,000 and less. So number one, don’t reduce—or excuse me, don’t raise taxes on middle-income people, lower them.

Hence, it’s unlikely that Romney would agree to eliminate all deductions for someone making between $100,000 and $200,000 per year. The table shows that eliminating all deductions for just those with incomes over $200,000 only provides about $82 billion in increased revenue. This is just 43 percent of the projected revenue loss of $190 billion. Hence, it would seem necessary that deductions would need to be reduced severely for workers making between $100,000 and $200,000 per year. In addition, the table above shows the key deductions that would need to be severely limited. They would be chiefly be the deductions for home mortgage interest, state and local taxes, real estate taxes, and charitable contributions.

Following are five points for Romney’s tax plan as given on his website:

• Make permanent, across-the-board 20 percent cut in marginal rates
• Maintain current tax rates on interest, dividends, and capital gains
• Eliminate taxes for taxpayers with AGI below $200,000 on interest, dividends, and capital gains
• Eliminate the Death Tax
• Repeal the Alternative Minimum Tax (AMT)

I think that the above table strongly suggests that these goals cannot be achieved without increasing the deficit, even if deductions are severely cut for those with incomes above $100,000. Of course, this analysis is not intended to be the final word on this subject. Its chief intention is to provide links to the key analyses and the IRS data that is being used in some of those analyses so that others can judge the issues for themselves.

Related articles

• The Romney Plan: Winners and Losers
• Even The Studies Paul Ryan Cited Prove Romney’s Tax Plan Is Impossible
• The 6 Studies Paul Ryan Cited Prove Mitt Romney’s Tax Plan Is Impossible
• The Final Word on Mitt Romney’s Tax Plan
• Ryan’s six studies backing up Romney’s tax plan turn out not to be six studies
• Bushonomics on Steroids: Romney’s Tax Plan Is Still Severely Impossible
• Princeton Economist: Romney Tax Plan Mathematically Sound
• Moody’s Chief Economist On Romney’s Tax Plan: ‘The Arithmetic Doesn’t Work’
• ‘Devoted’ Tax Specialist: Romney’s Tax Plan Is ‘Not Mathematically Possible’
• Another study tries to undermine the central premise of the Romney tax plan — and fails

The recent economic crisis left the median American family in 2010 with no more wealth than in the early 1990s, erasing almost two decades of accumulated prosperity, the Federal Reserve said Monday.

A hypothetical family richer than half the nation’s families and poorer than the other half had a net worth of $77,300 in 2010, compared with $126,400 in 2007, the Fed said. The crash of housing prices directly accounted for three-quarters of the loss.

The tables at this link are taken from the SCF and show that these numbers are in constant 2010 dollars. The graph to the right shows these inflation-corrected numbers since 1989. It’s clickable, so you can see it in full size.

As you can see, the top 10 percent has such a high relative net worth that the changes in the lower percentiles are difficult to discern. Hence, below is the same graph but going only up to a net worth of $240 thousand.

As you can see, median family net worth in 2010 was just about where it was in 1992, 18 years earlier. Hence, the economic crisis did erase “almost two decades of accumulated prosperity” as stated in the article for the median family. For the mean family, however, it just erased almost one decade of accumulated prosperity.

Table 4 of the 2010 SCF (from whence this data came) gives family net worth, by selected characteristics of families, from the past eight surveys (1989 through 2010). It gives both the mean and the median of the subgroups determined by these characteristics so it’s important to understand the difference between these two terms. Briefly, the mean of a series of numbers is the “average”, computed by dividing the sum of the numbers by the count of the numbers in the series. The median, on the other hand, is the “middle” value when the numbers are arranged in order of value. If there are an even number of values, it is the average of the two middle values. Hence, the series 1, 2, and 6 has a mean of 3 (9 divided by 3) and a median of 2 and the series 1, 2, 3, and 6 has a mean of 3 (12 divided by 4) and a median of 2.5. Note that in both of these cases, the mean is greater than the median. This is because the relatively high value of the last number in the series (the 6) pulls up the mean more than it does the median. This is important to remember in interpreting the numbers in table 4.

The first graph above shows the overall mean and median family net worth. Note that the overall mean is much higher than the median and is just below the median of the 75th to 90th percentile. This is because, like the example above, the majority of families have a net worth less than the mean. In fact, the above graph would indicate that about 80 percent of families have a net worth below the mean.

One useful thing about looking at the medians of percentiles is that the median represents the central percent of the percentile. That is, the medians of the 0 to 25, 25 to 50, 50 to 75, 75 to 90, and 90 to 100 percentiles represent the 12.5, 37.5, 62.5, 82.5, and 95 percentiles, respectively. That is because percentiles, like medians, are obtained by arranging the series by order of value. Hence, the data shows that 50 percent of families had a net worth less than $77,300 in 2010 and three-quarters of those families (37.5 percent of all families) had a net worth less than $32,200. This and the other data shown in the graphs above show that net worth is strongly skewed toward the upper percentiles.

The following table gives the data in the above graphs as displayed from the tables at this link:

Median Family Net Worth, by Selected Characteristics of Families

1989–2010 Surveys (thousands of 2010 dollars)

The last column shows the percent change net worth from 2007 to 2010. As you can see, the greatest percent loss to net worth was to the lower percentiles. The percent loss to the lowest 25 percent, second 25 percent, third 25 percent, next 15 percent, and top 10 percent were 100, 43.8, 31.9, 19.7, and 6.4 percent, respectively. Hence, the burden of the recent economic crisis weighed most heavily on those with the lowest net worth, at least in percentage terms.

Related articles

• Yes Paul Ryan is ‘Better off’: since 2008 he’s become a multimillionaire
• Economic Depression Devasted American Families’ Net Worth

Arthur Laffer isn’t a fan of Keynesian stimulus

On August 5th, The Wall Street Journal ran an editorial titled “Arthur Laffer: The Real ‘Stimulus’ Record”. The author, Arthur Laffer, is an American economist best known for the Laffer Curve. The editorial begins as follows:

Policy makers in Washington and other capitals around the world are debating whether to implement another round of stimulus spending to combat high unemployment and sputtering growth rates. But before they leap, they should take a good hard look at how that worked the first time around.

It worked miserably, as indicated by the table nearby, which shows increases in government spending from 2007 to 2009 and subsequent changes in GDP growth rates. Of the 34 Organization for Economic Cooperation and Development nations, those with the largest spending spurts from 2007 to 2009 saw the least growth in GDP rates before and after the stimulus.

Regarding this nearby table, monetary economist David Glasner says the following in a critique of the Laffer editorial:

So how did Laffer perform his calculation? He doesn’t say. All he does is cite the IMF as the source for his table. Thanks a lot, Art; that was really helpful, but unfortunately, not helpful enough to figure out what you are talking about.

Glasner is referring to the fact that the table gives the source of the data simply as “International Monetary Fund”. It does seem that many discussions in the political arena are hamstrung (perhaps purposely) by the failure of the author to give a usable source, much less to explain his calculations. I’ve therefore taken the time to do what Laffer arguably should have done to begin with. I found this spreadsheet on the IMF site which appears to contain the numbers used by Laffer. The following two tables show how these numbers can be used to calculate Laffer’s figures:

Gross Domestic Product, Constant Prices
Percent change

Country	2006	2007	2008	2009	2010	2011	(08+09) minus (06+07)
United States	2.658	1.913	–0.337	–3.486	3.030	1.735	–8.4
Japan	1.693	2.192	–1.042	–5.527	4.435	–0.748	–10.5
Germany	3.889	3.394	0.809	–5.078	3.562	3.056	–11.6
France	2.658	2.234	–0.196	–2.631	1.382	1.715	–7.7
United Kingdom	2.607	3.466	–1.103	–4.373	2.092	0.655	–11.5
Italy	2.199	1.683	–1.156	–5.494	1.804	0.431	–10.5
Canada	2.823	2.200	0.689	–2.770	3.215	2.460	–7.1
Australia	2.682	4.676	2.500	1.373	2.544	2.035	–3.5
Spain	4.077	3.479	0.888	–3.740	–0.070	0.710	–10.4
Mexico	5.147	3.242	1.186	–6.275	5.543	3.967	–13.5
Korea	5.179	5.106	2.298	0.319	6.320	3.634	–7.7
Turkey	6.893	4.669	0.659	–4.826	9.006	8.460	–15.7
Netherlands	3.394	3.921	1.804	–3.479	1.633	1.266	–9.0
Switzerland	3.630	3.645	2.095	–1.878	2.714	1.851	–7.1
Sweden	4.557	3.431	–0.774	–4.845	5.845	3.991	–13.6
Poland	6.227	6.785	5.127	1.606	3.944	4.350	–6.3
Norway	2.443	2.652	0.009	–1.661	0.654	1.688	–6.7
Belgium	2.702	2.900	0.957	–2.841	2.266	1.893	–7.5
Austria	3.670	3.706	1.396	–3.810	2.315	3.107	–9.8
Denmark	3.395	1.583	–0.784	–5.834	1.296	1.050	–11.6
Chile	5.825	5.207	3.034	–0.860	6.137	5.924	–8.9
Greece	4.614	3.032	–0.137	–3.258	–3.507	–6.860	–11.0
Finland	4.411	5.335	0.294	–8.354	3.732	2.855	–17.8
Israel	5.594	5.497	4.028	0.837	4.846	4.707	–6.2
Portugal	1.448	2.365	–0.008	–2.908	1.383	–1.466	–6.7
Ireland	5.312	5.182	–2.972	–6.995	–0.430	0.705	–20.5
Czech Republic	7.020	5.735	3.099	–4.695	2.739	1.655	–14.4
New Zealand	0.997	2.840	–0.074	–2.071	1.215	1.441	–6.0
Hungary	3.900	0.100	0.900	–6.800	1.270	1.695	–9.9
Slovak Republic	8.345	10.494	5.751	–4.932	4.183	3.349	–18.0
Luxembourg	4.969	6.639	0.754	–5.300	2.678	1.004	–16.2
Slovenia	5.850	6.870	3.589	–8.008	1.380	–0.175	–17.1
Estonia	10.097	7.492	–3.671	–14.258	2.264	7.636	–35.5
Iceland	4.709	5.985	1.270	–6.807	–4.024	3.051	–16.2

General Government Total Expenditure
Percent of GDP

Country	2006	2007	2008	2009	2010	2011	2009 minus 2007
United States	35.852	36.672	39.196	43.981	42.142	41.397	7.3
Japan	34.489	33.312	35.730	39.982	39.002	40.675	6.7
Germany	45.559	43.506	44.046	48.104	47.869	45.625	4.6
France	52.934	52.595	53.329	56.725	56.668	56.321	4.1
United Kingdom	40.613	40.325	43.066	47.267	46.325	45.725	6.9
Italy	48.464	47.611	48.610	51.889	50.495	49.953	4.3
Canada	39.268	39.158	39.535	44.056	43.815	42.664	4.9
Australia	34.635	34.244	34.487	37.579	36.773	36.598	3.3
Spain	38.342	39.192	41.300	46.065	45.439	43.586	6.9
Mexico	22.818	23.148	24.606	28.314	26.928	26.211	5.2
Korea	21.535	21.886	22.387	23.030	20.999	21.658	1.1
Turkey	32.793	33.325	33.837	37.717	35.443	34.186	4.4
Netherlands	45.687	45.087	46.116	50.785	50.614	50.034	5.7
Switzerland	35.663	34.634	32.612	34.398	34.040	34.736	–0.2
Sweden	50.770	48.968	49.625	52.793	50.634	49.126	3.8
Poland	43.864	42.187	43.194	44.510	45.366	44.472	2.3
Norway	39.936	40.386	39.764	46.610	45.421	44.320	6.2
Belgium	48.603	48.320	49.900	53.828	52.942	53.450	5.5
Austria	49.141	48.602	49.335	52.893	52.603	50.449	4.3
Denmark	51.250	50.788	51.424	57.899	56.174	55.972	7.1
Chile	18.716	19.381	21.727	24.622	23.628	23.300	5.2
Greece	44.691	46.709	49.734	53.033	49.620	49.696	6.3
Finland	49.238	47.431	49.316	56.103	55.550	54.043	8.7
Israel	47.501	46.016	45.427	45.086	44.736	44.354	–0.9
Portugal	44.364	44.362	44.816	49.914	51.414	48.710	5.6
Ireland	33.406	36.194	42.301	47.937	65.637	44.143	11.7
Czech Republic	41.967	41.040	41.148	44.921	44.110	44.549	3.9
New Zealand	31.105	31.064	32.862	34.471	34.490	35.364	3.4
Hungary	52.164	50.641	49.205	51.380	49.453	48.441	0.7
Slovak Republic	36.521	34.210	35.049	41.704	41.100	38.375	7.5
Luxembourg	38.576	36.267	37.103	43.042	42.482	41.635	6.8
Slovenia	42.547	40.262	41.422	46.355	47.134	47.713	6.1
Estonia	34.573	34.853	41.037	47.654	44.713	43.123	12.8
Iceland	41.641	42.268	44.639	49.672	47.923	46.321	7.4

The numbers in the rightmost column of both tables exactly match the numbers in Laffer’s table. Hence, the first table shows that the change in real GDP growth in Laffer’s table is equal to the annual GDP changes in 2008 and 2009 minus the annual GDP changes in 2006 and 2007. The second table shows that the change in government spending in Laffer’s table is equal to the government spending in 2009 minus the government spending in 2007, both as a percent of GDP.

Based on these numbers, his editorial continues as follows:

The four nations—Estonia, Ireland, the Slovak Republic and Finland—with the biggest stimulus programs had the steepest declines in growth. The United States was no different, with greater spending (up 7.3%) followed by far lower growth rates (down 8.4%).

This statement is based on his calculations for these four nations, shown in bold in the rightmost column of the above tables. Regarding Laffer’s calculation of public spending as a percent of GDP, economist Lars Christensen says the following in another critique of the Laffer editorial:

One major problem with Laffer’s numbers is that he is using public spending as share of GDP to analyze the magnitude of change in fiscal policy. However, for a given level of public spending in euro (the currency today in Estonia) a drop in nominal GDP will naturally lead to an increase in public spending as share of GDP. This is obviously not fiscal stimulus. Instead it makes more sense to look at the level of public spending adjusted for inflation and this is exactly what I have done in the graph below. I also plot Estonian GDP growth in the graph. The data is [sic] yearly data and the source is IMF.

GDP Growth And Government Expenditures for Estonia: 2006–2011

To the right is a similar graph, also showing public spending adjusted for inflation (blue line) and GDP growth (red line) for Estonia. It also shows public spending as a percentage of GDP (green line), the spending values used by Laffer.

The following table shows the values for real public spending as calculated from the IMF data:

General government total expenditure
Constant national currency* (index 2006=100)

Country	2006	2007	2008	2009	2010	2011	2009 minus 2007
United States	100.0	104.3	109.4	120.1	118.0	116.7	15.8
Japan	100.0	97.7	101.0	107.7	108.2	110.1	10.0
Germany	100.0	98.0	98.3	102.9	105.6	101.9	4.9
France	100.0	102.6	103.2	107.2	107.6	108.0	4.7
United Kingdom	100.0	102.7	107.9	112.7	112.3	109.3	10.1
Italy	100.0	100.2	100.2	102.4	100.2	98.0	2.2
Canada	100.0	103.0	106.4	112.8	117.1	117.2	9.8
Australia	100.0	105.5	111.3	121.1	124.5	127.6	15.6
Spain	100.0	106.2	111.0	119.6	116.0	110.2	13.4
Mexico	100.0	106.4	115.8	123.8	124.2	128.2	17.4
Korea	100.0	106.3	109.4	113.6	110.9	115.9	7.3
Turkey	100.0	103.9	107.7	113.2	113.5	121.5	9.3
Netherlands	100.0	102.8	107.0	112.2	114.0	112.9	9.3
Switzerland	100.0	102.4	98.5	102.6	103.6	108.2	0.1
Sweden	100.0	100.7	101.3	102.4	103.3	103.8	1.7
Poland	100.0	104.2	110.9	116.4	121.9	123.3	12.2
Norway	100.0	106.2	111.8	118.1	120.3	124.5	11.9
Belgium	100.0	102.8	104.8	111.2	111.3	113.0	8.4
Austria	100.0	102.4	103.9	107.8	109.8	107.0	5.4
Denmark	100.0	101.2	102.5	108.4	108.1	106.9	7.1
Chile	100.0	109.3	116.9	134.6	145.3	151.1	25.3
Greece	100.0	109.1	116.2	121.3	106.2	97.6	12.3
Finland	100.0	102.9	106.3	110.5	112.1	112.5	7.7
Israel	100.0	102.1	101.5	103.3	105.9	108.5	1.2
Portugal	100.0	102.8	102.7	113.1	117.7	107.1	10.4
Ireland	100.0	112.2	120.6	124.0	167.5	111.7	11.7
Czech Republic	100.0	103.9	102.9	108.0	105.6	105.5	4.2
New Zealand	100.0	104.9	110.0	114.0	117.2	121.1	9.1
Hungary	100.0	94.9	92.4	89.4	85.6	85.0	–5.6
Slovak Republic	100.0	102.7	110.1	122.0	125.0	117.8	19.3
Luxembourg	100.0	101.6	105.7	115.8	120.4	118.9	14.3
Slovenia	100.0	101.7	106.8	112.2	112.4	112.4	10.6
Estonia	100.0	113.5	122.8	121.2	114.2	117.0	7.7
Iceland	100.0	108.2	115.2	115.4	108.4	107.1	7.2

A careful comparison of Christensen’s graph and the graph above shows that real public spending (blue line) increases slightly faster on the former, reaching about 122 in 2011 versus 117 in the latter. This is likely because the graph above is using “Inflation, average consumer prices” as given in the IMF data, whereas Christensen’s data is likely using an alternate measure of inflation. However, the shapes of the line are basically the same and Christensen’s following statement holds true for both:

So what happened in 2009? Inflation adjusted public spending dropped! This is what makes Estonia unique. The Estonian government did NOT implement Keynesian policies rather it did the opposite. It cut spending. This is clear from the graph (the blue line). It is also clear from the graph that the Estonian government introduced further austerity measures and cut public spending further in 2010. This is of course what Laffer calls “fiscal stimulus”.

In fact, the four countries with the most fiscal stimulus as measured by the real increase in public spending from 2007 to 2009 were Chile (25.3), the Slovak Republic (19.3), Mexico (17.4), and the United States (15.8)! That includes only one of Laffer’s choice of four countries with the lowest growth in GDP rates. The other three had much lower increases in public spending of 11.7 (Ireland) and 7.7 (Estonia and Finland).

In summary, Laffer makes a number of mistakes in his editorial. The first is to compare growth in GDP rates with government spending as a percent of GDP. He is testing for a relationship between two variables but expressing one of them (spending) in terms of the other (GDP). That is, he is linking them by design! As an example of this link, suppose that spending remains constant but GDP drops. By simple arithmetic, this will cause spending as a percentage of GDP to rise and be interpreted by Laffer as stimulus. This seems like an incredibly elementary mistake for a professional economist to be making.

The second mistake is to select one span of data and look at only those data. Laffer chooses to look at GDP growth from 2006 to 2009 and government spending from 2007 to 2009. The tables at this link show calculations for these same variables going out to 2011 instead of 2009. As is evident, this causes the results to be very different. For example, the four nations with the highest drop in GDP growth rates are Slovenia (-23.1), the Slovak Republic (-22.0), Greece (-21.1), and Iceland (-19.7), when the span is increased from 2009 to 2011.

It’s often a good idea to look at all of the available data in graphical format (such as the one above) to get a look over all possible time spans. Similar graphs for Finland, Ireland, and the Slovak Republic can be found at the aforementioned link.

Finally, Laffer makes the mistake of not giving a precise source that allows his calculations to be checked. This may be as much of a mistake of The Wall Street Journal as it is for Laffer. I believe that all publications should mandate that precise sources be given and, if possible, links be provided to background material that explains the author’s calculations. Then, others will be able to check the author’s work, especially in the case where the publications chooses not to do so or does a poor job of doing so.

Related articles

• Arthur Laffer’s Anti-Stimulus Curve Ball is a Foul (business.time.com)
• Arthur Laffer: The Real ‘Stimulus’ Record — WSJ.com (cg68doc.newsvine.com)
• Arthur Laffer: Wall Street Journal Op-Ed Proves He Still Has Much to Learn (thenewamerican.com)