Back then, $1 billion was big money. Yet, the Military Aviation Appropriations Act of 1917 and two subsequent bills passed with almost no debate. It was virtually a blank check to buy an air armada to fight World War I.

The New York Times hailed the legislation, predicting that America would have "such an air fleet as the world has never known." The battlefield skies would be flooded with thousands of American planes.

It didn't work out that way. By July 1918, there was still not one U.S.-built plane flying in combat. In the end, the military got fewer than 200 planes for the $1 billion.

But Gen. "Black Jack" Pershing, commander of U.S. forces in Europe, did a little better. Notes Linda Robertson in "The Dream of Civilized Warfare," he managed to buy 5,000 planes and 8,500 engines "off the shelf" from the French ... for $60 million.

In retrospect, it's easy to see what went wrong. America simply did not have an industrial base suited to cranking out that many airplanes that fast. Congress probably should have thought of that before it ponied up a billion in taxpayers' money.

Today, our government has gone to the other extreme. Military procurement is now so risk averse, with so many checks and balances, that tax dollars now buy mostly red tape.

Case in point: the Army's research, development, test and evaluation program. Before 2004, it spent, on average, about $1 billion annually on programs that wound up being canceled. Today, that figure is four times higher. This year, the program is left holding a $4 billion bag of nothing.

The next time Defense Secretary Robert Gates trumpets how he has "saved" money by canceling programs -- Congress ought to look more closely. Gates' idea of "savings" is like making a $50,000 down payment on a $500,000 house, walking away from the deal and claiming to have saved $450,000.

This is not just an Army problem. Some in Congress want to put the Gates saving plan on steroids. The Pentagon has spent about $50 billion developing the F-35, a state-of-the-art fighter plane.

Congress keeps cutting back the number of planes it plans to buy. The smaller quantities drive up the per-unit cost. Worse, when you pro-rate the upfront price of research and add it to the cost of each plane -- the price tag is astronomical. It would be like Steve Jobs developing the iPad and producing only 10 of them -- and tacking all the research costs on those 10 iPads.

Retired Gen. Dave Maddox recently sat on a commission that reviewed the Army's acquisition program from top to bottom. What he found was a shoot-self-in-foot process. His conclusion: "We don't need an enemy."

The commission report denounced the program's risk-averse system and culture. Instead of "managing risk," the Rube Goldberg process set up by legislative dictates and Pentagon bureaucrats has become all about "avoiding risk."

To give taxpayers value for money, Congress and the Pentagon must start actually buying military equipment rather than just spend billions and buy nothing.

Unless America gets back to buying military ships, planes and vehicles soon, the industries that provide these goods will dry up and blow away. Then, when the next war comes, we'll be back to 1917: anxious to spend billions, but no place to spend them.

Well, maybe the Chinese will sell us something.