Hey Big Spender

Meet the new boss
Same as the old boss
--The Who

Once a week, John Stossel hosts a weekly talk show on FoxBusiness that, near as I can tell, is the only program discussing today's issues from a libertarian perspective.

This week his topic was government spending. During the show, he questioned Republican claims to shrink government based on a historical record that does not match party rhetoric.

He grilled Karl Rovel about the gigantic spending record of the Bush administration. Rove's answers were pathetic and I wish Stossel would have pressed him more.

Using the same Office of Management and Budget source as Stossel, let's examine the data a bit more closely. The chart below shows federal government outlays since 1949. This year was the beginning of Truman's second term and represented the first full presidential term post WWII. This seemed the a reasonable starting point for 'modern' times.

Outlays under Democrat adminstrations are indicated in blue while outlays under Republican administrations are indicated in red. Visual inspection alone suggests little difference in spending habits between regimes. Spending goes up regardless of who's in office (although there were actually three declines in outlays during the above period--two under Ike and one under LBJ).

Moreover, rate of change in outlays does not appear visually different either (the curve increases smoothly over time). One might argue that rate of change increased a bit less under Clinton. But Bush picked the pace back up and Obama has steepened the geometric progression.

From a statistical standpoint, the total period above represents 62 years (counting this year) with an average year over year change in federal government outlays of +8.4%. (GDP increased +6.8% annually during this period)

Republican administrations dominate the time period 36 yrs to 26. Average annual percentage change in outlays during Democrat administrations is higher than that during Republican regimes: 10.2% annually for Dems vs 7.0% for GOP. However, a simple two sample t-test suggests that the difference in means is not significant (p = .16).

Both the eyeball and the stats tell us the same thing. In general, Democrats and Republicans display similar spending habits.