Those who hoped that Jeff Bezos takeover of the Washington Post would lead to a quick improvement in the quality of its budget reporting will be seriously disappointed by the paper's lead story today. The story bemoaned the fact that, "after six budget showdowns, big government is mostly unchanged" [the article's headline].

The article uses four metrics to measure the size of government, none of which would inform readers of anything. Its lead metric is spending in nominal dollars, which it tells us will be $3.455 trillion in fiscal 2013. It tells us that this is down by only a small amount from a "whopping $3.457 trillion" spent in 2010.

Incredibly, the article does not even adjust this spending amount for inflation. (The piece does briefly note later that this is a 5 percent decline adjusted for inflation.) Of course a serious analysis would have expressed spending as a share of GDP, which shows that spending dropped from 24.1 percent of GDP in 2010 to 21.5 percent of GDP in 2013. This decline in spending of 2.6 percentage points of GDP would be the equivalent of roughly $420 billion in today's economy.

Assuming a multiplier of 1.5, this reduction in spending has cost the economy more than $600 billion in annual output since there is no plausible story by which cuts in government spending lead to addition private sector demand in the current economic situation. (To be fair, there is a lot of vigorous handwaving on this topic by proponents of spending cuts.) That would translate into more than 5 million fewer jobs.

The piece goes on to tell us that Bezos' paper does not like government spending in general and in particular dislikes Social Security and Medicare. In terms of government spending the piece tells readers:

"But even now, the government still spends a vast amount of money.

"This year’s projected spending will be more than in any year of the George W. Bush administration. And more than 30 percent higher (accounting for inflation) than the last year of President Clinton’s term."

Those who are interested in serious historical analysis would note that government spending as a share of GDP is lower in 2013 than it was in every year of the Reagan presidency except 1988 when spending was 21.3 percent of GDP, 0.2 percentage points lower than the 2013 share. Using the normal metric of spending as a share of GDP means that the Post's reference to a:

"decades-long spending binge that peaked in the first years of President Obama’s term"

is simply untrue. In fact, there has been no consistent upward trend in government spending as a share of GDP.

The Post lets us know of its dislike for Social Security and Medicare by telling us that Congress "failed" to deal with these programs. By "deal" the Post means cut. The reason that Congress did not cut these programs is that the people across the political spectrum overwhelmingly support these programs. In fact, more people would rather see these programs expanded than cut. While it may disappoint Bezos and his employees at the Post, it should not be surprising that elected officials would not vote to cut programs that most of their constituents support.

The Post's other metrics of the size of government are just silly. It tries to give us numbers of government employees. [The article perversely puts the number of employees at 4 million. Those used to Labor Department data might be more familiar with the number of employees as 2,160,000 or 2,750,000 if we include the Post Office. The 4 million number includes uniform military personnel.] As the article notes, this doesn't really mean anything since often employees are replaced by contract workers. The article then expresses its unhappiness that the government has not spent more money trying to find out exactly how many people work on government contracts:

"It’s hard to judge the actual size of the government — or the actual scope of its work — without knowing how many of these people [contract workers] exist.

"The Obama administration doesn’t. It was supposed to have started counting these contractors: Congress ordered it in 2009. But the formal regulations haven’t been finalized. So there is still no full count. There are only educated guesses."

It actually is difficult to define a government contract worker. Is every person who gets a contract for work with the government (e.g. a researcher getting a $10,000 grant for a project) a contract worker? If the government replaces lawn care workers with a lawn service, should these workers be considered contract employees? Would we count the workers at Lockheed who design and manufacture weapon systems for the government as contract employees?

There are not clear lines and it's not obvious what we find out from this exercise. But apparently the Post wants the government to spend lots of money doing this calculation.

Then we have pages of federal regulations. This is an extremely perverse measure of the size of government since often pages are added to reduce government authority. For example, the Dodd-Frank bill got considerably longer as the industry wrote in large areas of exemptions from various rules.

Finally, we have the number of buildings rented or owned by the government, which the Post describes as "staggering." (Maybe it should find budget reporters who can work with big numbers without falling off their feet.) Since the government can easily contract out work, it can find simple ways to reduce office space if that mattered for anything. It is also not clear what the "building" measure is supposed to mean. If the government replaces offices in two small buildings with offices in one big building is that a victory for small government?

So that you have it folks: budget reporting at the Post in the Bezos era, more biased and confused than ever. Makes you wonder if Amazon ever would have survived without the massive subsidy provided by being exempted from the requirement to collect sales tax.