The Vast Left-Wing Wonk Machine

Fact-checkers are at it again.  They are calling it a myth that ObamaCare is the biggest tax increase in history.  But let's see how these great fact-checkers work.

Bloomberg Businessweek has an article titled "Why Obamacare's Tax Increase Isn't the Biggest Ever."  Its source: "a nice chart from Ezra Klein."  Let's go over to Ezra.

Ezra Klein got that nice chart from Austin Frakt, an economist who blogs.  And Austin Frakt got the data for that chart from Kevin Drum.  And Kevin Drum is a writer for (drum roll) the left-wing magazine Mother Jones.  His tone and objectivity come across in this sentence: "This is so stupid it hurts."

So far we have Bloomberg and the Washington Post simply repeating something that Mother Jones originated.

But at least Drum was using something almost credible: a working paper from the Treasury department.  The paper Drum cited provided only tables of results, with no explanation of methodology.  As it turns out, the description of that methodology is given in a previous version of that working paper, here.

Now we are starting to get somewhere.  The author of that Treasury paper, Jerry Tempalski, got his numbers from Treasury estimates done years ago, when the legislation occurred.  But let me provide an excerpt from Tempalski's paper. 

Treasury revenue estimates presented in the Administration's first post-enactment January budget were used. Treasury produces estimates for a bill when it is enacted, generally reestimates the bill for the January budget, and sometimes reestimates a bill for several subsequent budgets. In a few cases, the first post-enactment estimates proved not very accurate. For example, revenues from the Crude Oil Windfall Profit Tax of 1980 were estimated to be $21 billion in 1984; actual revenues were probably less than $4 billion. This paper makes no adjustment for estimates that proved to be inaccurate. [My emphasis.]

In other words, all the charts and tables you saw spread throughout Mother Jones, the Washington Post, Bloomberg and even Politifact came from Treasury predictions that were done before any actual revenue data were measured!  Even Tempalski, the original author of all this, showed that such predictions could be off by over 400%!

Let's sanity-check those predictions, shall we?  Tempalski's tally had 22 revenue bills from 1968 through 2006.  Add up all those predictions as changes to GDP (using the first 2-year average, just as Drum, and therefore Frakt, Klein, and Bloomberg, did) and the overall net result is minus 2.4% of GDP.  The actual change in revenue from 1968 to 2007 was plus 0.9% of GDP.  Below is my attempt to show how these predictions matched actual revenue over time.

 

Sources: Actual data from OMB.  Predicted data are the author's calculations based on Treasury's first-2-year-average predictions, accumulated over time from the actual revenue in the initial year of 1968.  Changes in predictions occur in the year the given revenue bill was enacted.

The total error of these 22 predictions was over $450 billion in 2007.  Not only was the error large, but in the wrong direction: predictions said revenues would fall, but they actually rose.  You would have been more correct than Treasury's predictions if you had simply estimated that revenue bills have zero impact on revenue!

Treasury's predictions use static scoring, meaning Treasury assumes that changes in tax rates have zero effect on consumer and investor behavior.  Yes, that is as stupid as it sounds, and we see the result: a total inability to predict revenues with any degree of accuracy.

By the way, I am not criticizing Tempalski.  He had to use something, and Treasury predictions were fairly well-documented and internally consistent.  It just happens that Treasury predictions are useless.  You can tabulate and graph useless predictions to your heart's content, and publish them in a hundred newspapers.  But they are still useless predictions.  And in this case, the years-old predictions had already been invalidated by measured data!  It's models all the way down.

As for ObamaCare (or the ACA, if you will), one of two trustees of Social Security and Medicare, appointed by President Obama, estimated the following.

Over the years 2012-21, the ACA is expected to add at least $340 billion and as much as $530 billion to federal deficits while increasing federal spending by more than $1.15 trillion over the same period and by increasing amounts thereafter.

I have to admit, that would not be the biggest deficit bill ever.  Obama's Stimulus has that beat by a cool quarter-trillion at least.  Not to mention, any offset from that $1.15 trillion in spending comes from cuts to Medicare and from new taxes and penalties.  It's still a trillion-plus spending bill.

Follow the data trail: useless Treasury predictions, a Treasury report compiling those predictions, Mother Jones and Politicfact glomming onto that report, an economist blogger graphing that data, Ezra Klein reporting that graph in the Washington Post, and Bloomberg citing Ezra Klein to debunk a "myth."

It's almost like the way Hillary Clinton described a certain vast conspiracy.

Randall Hoven can be followed on Twitter.

Fact-checkers are at it again.  They are calling it a myth that ObamaCare is the biggest tax increase in history.  But let's see how these great fact-checkers work.

Bloomberg Businessweek has an article titled "Why Obamacare's Tax Increase Isn't the Biggest Ever."  Its source: "a nice chart from Ezra Klein."  Let's go over to Ezra.

Ezra Klein got that nice chart from Austin Frakt, an economist who blogs.  And Austin Frakt got the data for that chart from Kevin Drum.  And Kevin Drum is a writer for (drum roll) the left-wing magazine Mother Jones.  His tone and objectivity come across in this sentence: "This is so stupid it hurts."

So far we have Bloomberg and the Washington Post simply repeating something that Mother Jones originated.

But at least Drum was using something almost credible: a working paper from the Treasury department.  The paper Drum cited provided only tables of results, with no explanation of methodology.  As it turns out, the description of that methodology is given in a previous version of that working paper, here.

Now we are starting to get somewhere.  The author of that Treasury paper, Jerry Tempalski, got his numbers from Treasury estimates done years ago, when the legislation occurred.  But let me provide an excerpt from Tempalski's paper. 

Treasury revenue estimates presented in the Administration's first post-enactment January budget were used. Treasury produces estimates for a bill when it is enacted, generally reestimates the bill for the January budget, and sometimes reestimates a bill for several subsequent budgets. In a few cases, the first post-enactment estimates proved not very accurate. For example, revenues from the Crude Oil Windfall Profit Tax of 1980 were estimated to be $21 billion in 1984; actual revenues were probably less than $4 billion. This paper makes no adjustment for estimates that proved to be inaccurate. [My emphasis.]

In other words, all the charts and tables you saw spread throughout Mother Jones, the Washington Post, Bloomberg and even Politifact came from Treasury predictions that were done before any actual revenue data were measured!  Even Tempalski, the original author of all this, showed that such predictions could be off by over 400%!

Let's sanity-check those predictions, shall we?  Tempalski's tally had 22 revenue bills from 1968 through 2006.  Add up all those predictions as changes to GDP (using the first 2-year average, just as Drum, and therefore Frakt, Klein, and Bloomberg, did) and the overall net result is minus 2.4% of GDP.  The actual change in revenue from 1968 to 2007 was plus 0.9% of GDP.  Below is my attempt to show how these predictions matched actual revenue over time.

 

Sources: Actual data from OMB.  Predicted data are the author's calculations based on Treasury's first-2-year-average predictions, accumulated over time from the actual revenue in the initial year of 1968.  Changes in predictions occur in the year the given revenue bill was enacted.

The total error of these 22 predictions was over $450 billion in 2007.  Not only was the error large, but in the wrong direction: predictions said revenues would fall, but they actually rose.  You would have been more correct than Treasury's predictions if you had simply estimated that revenue bills have zero impact on revenue!

Treasury's predictions use static scoring, meaning Treasury assumes that changes in tax rates have zero effect on consumer and investor behavior.  Yes, that is as stupid as it sounds, and we see the result: a total inability to predict revenues with any degree of accuracy.

By the way, I am not criticizing Tempalski.  He had to use something, and Treasury predictions were fairly well-documented and internally consistent.  It just happens that Treasury predictions are useless.  You can tabulate and graph useless predictions to your heart's content, and publish them in a hundred newspapers.  But they are still useless predictions.  And in this case, the years-old predictions had already been invalidated by measured data!  It's models all the way down.

As for ObamaCare (or the ACA, if you will), one of two trustees of Social Security and Medicare, appointed by President Obama, estimated the following.

Over the years 2012-21, the ACA is expected to add at least $340 billion and as much as $530 billion to federal deficits while increasing federal spending by more than $1.15 trillion over the same period and by increasing amounts thereafter.

I have to admit, that would not be the biggest deficit bill ever.  Obama's Stimulus has that beat by a cool quarter-trillion at least.  Not to mention, any offset from that $1.15 trillion in spending comes from cuts to Medicare and from new taxes and penalties.  It's still a trillion-plus spending bill.

Follow the data trail: useless Treasury predictions, a Treasury report compiling those predictions, Mother Jones and Politicfact glomming onto that report, an economist blogger graphing that data, Ezra Klein reporting that graph in the Washington Post, and Bloomberg citing Ezra Klein to debunk a "myth."

It's almost like the way Hillary Clinton described a certain vast conspiracy.

Randall Hoven can be followed on Twitter.

RECENT VIDEOS