David R. Henderson  

U.S. Federal Budget Cuts in the 1990s

Eubulides, Wilkinson, and Disc... Gains from Exchange...

The Untold Story that Needs to Be Told

I've referred on this blog a number of times to "Canada's Budget Triumph," my 2010 study for Mercatus on the successful budget cuts the Canadian government undertook from 1994 to the early 2000s. Here's one line from my study that sums up one of the key messages:

The result of years of cuts in government spending was that, as a percent of GDP, federal spending on programs [as distinct from interest on the federal debt] fell from a high of 17.5 percent in 1992-93 to 11.3 percent in 2000-01.

When I briefed this on Capitol Hill and when I've spoken about this in interviews, the push-back I've often got is that the United States can't do the same thing because Canada has a parliamentary system in which the executive branch and the legislative branches are one whereas the United States has a system in which the executive and legislative branches are separate. I agree that it's harder to do in the United States. But the evidence that it can be done is right under our noses: it's called the 1990s.

In 1990, total federal government spending, including interest on the federal debt, was 21.8 percent of GDP. By 2000, it was down to 18.4 percent of GDP, a 3.4-percentage-point drop. How did that happen? There were three factors:

1. Bush I and Clinton on military spending. Bush I, noticing that the Cold War had ended, started implementing cuts in military spending as a % of GDP. When Clinton came into office in January 1993, he kept on the glide path that Bush I had put in place. Result: Defense spending actually fell in nominal terms, from $299.3 billion in 1990 to $294.5 billion in 2000. As a % of GDP, it fell from 5.2 to 3.0, accounting for 2/3 of the drop.

2. The virtuous circle on net interest on the debt. As economic growth and restraint on program spending (federal spending that is not interest on the debt) kicked in, net federal debt peaked in 1993 at 49.5 percent of GDP and then fell to 35.1% of GDP by 2000. With a lower debt/GDP, interest payments on the debt were lower as a % of GDP, making the debt to GDP shrink further. Also helping were falling interest rates through the 1990s. Of course, with interest rates at or close to historical lows, we can't rely on that now. Net interest on the federal debt fell from 3.2% of GDP in 1990 to 2.3% of GDP in 2000, a fall of 0.9 percentage points.

3. Domestic spending: the dog that didn't bark. With the Republican Congressional victories in the November 1994 elections, Republicans took over both the U.S. House of Representatives and the U.S. Senate. Newt Gingrich, as the Speaker of the House, pushed hard for restraint in the growth of domestic spending. He got it. Domestic spending (which I define here as overall federal spending minus national defense minus international affairs minus net interest) fell very slightly from 13.0% of GDP to 12.9% of GDP.

Of course, the two reasons a fraction will fall are that the numerator falls and that the denominator increases. We had a boom in the 1990s. That's the denominator growing. But isn't that interesting? With many hundreds of thousands of jobs lost in the Department of Defense (military and civilian) in the defense-related industries, we had a boom.

Comments and Sharing

CATEGORIES: Fiscal Policy

COMMENTS (10 to date)
Sol writes:

Maybe I'm just becoming a crank in my middle age, but I've become very suspicious of articles that talk about budget solely in terms of percentage of GDP. It seems like it obscures what is really going on with the budget by confusing it with how the overall economy is doing.

In 1992, GDP was something like $6,500 billion. In 2000, it was about $10,000 billion. Therefore total federal government spending was about $1400 billion in 1992, and $1800 billion in 2000. So spending went up 29% in that time period; GDP just went up faster.

No idea if those figures are adjusted for inflation.

Jonathan Bechtel writes:


I was thinking the same thing. I probably agree with David's overall point, but more and more I think the macroeconomic events of the 1990's were due to serendipity more than anything else.

Most of Clinton's "progressive" measures were shot down because his party got booted out of Congress, and the tech/stock/housing boom was a rising tide that lifted all boats.

Most stories about the political direction the country took are backwards looking and retrofit events to convenient stories.

Mercer writes:

I don't think it is realistic to think domestic spending will fall when the number of seniors is going up. Even Ryan's plan would to start to lower Medicare spending only after ten years.

marcus nunes writes:

China has the second largest economy. That´s just because it has more people than Europe North and South America and Australia/N Zealand.
What matters to measure how rich a country is is income per capita. Under that measure, China has 1/5 of US income.
The same with government spending or receipts. You have to scale it to the size of the economy. Population grows and gets old, expenditures with SS, medicare etc go up. But government expenditures overall fell relative to income exactly because they didn´t grow as fast as income (GDP).

MingoV writes:

I agree with Sol: there is no reason why federal government spending should move in parallel with the GDP. For year-to-year federal spending comparisons the only two modifiers should be inflation and the change in the number of people receiving Social Security and Medicare benefits.

Federal spending (in 2005 dollars) increased by $380 billion from 1993 ($1,409 billion) to 2000 ($1,789 billion). $170 billion of that increase was due to increased spending on Social Security and Medicare (a combination of more beneficiaries and more spending per beneficiary). Annual federal spending (corrected for inflation and increased Social Security and Medicare spending) increased by an average of 1.5% from 1993 to 2000 with a total increase of 15%. During that same period, the annual GDP (corrected for inflation) grew by an average of 3.3% per year with a total increase of 26%. That large GDP increase hides the increased federal spending if one looks at the ratio of spending to GDP.

Sol writes:

marcus, of course "But government expenditures overall fell relative to income exactly because they didn´t grow as fast as income (GDP)." But that does nothing to tell us whether this is because of expenditures or income!

A fictionalized example: My family's spending relative to income is much lower this year than last. This post is doing the equivalent of crediting that to our improved spending habits, while ignoring that my wife got a job this year.

David R. Henderson writes:

This post is doing the equivalent of crediting that to our improved spending habits, while ignoring that my wife got a job this year.
No, it’s not. Read the last paragraph of my post.

Ken B writes:

Part of the arguemtn is that government spending is usually misallocated. So if the resources tied up doing government work are freed to do other, more productive, work then total production will rise as a result.

If indeed most govt spending IS misallocated then govt size is a decent measure of that misallocation. That suggests that fraction of GDP is a reasoanble first-order measure of extent of the misallocation.

Mr. Econotarian writes:

The specifics of Canada's 1990's budget cuts are the most interesting...

"a reduction in unemployment insurance benefits brought about by reducing the duration of benefits" - the US has done the exact opposite

"additional cuts in defense spending totaling $1.9 billion over the next years" - yet the US has done little to reduce defense spending, and indeed is involved in new wars of questionable ROI

"cuts in subsidies for businesses" - The US has had various stimulus efforts for Solyndra, cellulosic ethanol, high speed rail, etc.

Ken B writes:

@ Me Econotrarian:

Shhh. Don't let the cat out the bag. We have a reputation as nice, solicitous neighbours. But our real plan is to buy your country up at firesale prices when you default. That's why we got our house in order.

Jesting aside, the Obama 'stimulus' is seen by most Canadians as a good thing, judging from my experience, and most Canadians strongly favour the democrats. This is mostly a cultural prejudice thing I think not a reasoned analysis. But how quickly we forget.

Comments for this entry have been closed
Return to top