Yesterday I got to make fun of the NYT for telling readers that food stamps was a $760 billion program. I thought they were giving the 10-year cost, but apparently they had just made a mistake and added a zero to the annual cost, as they later acknowledged in a correction.

This helped make my point. No one, apparently including the editors at the NYT, has any idea what these budget numbers mean. If the original article had told readers that spending on the food stamp program was roughly 1.8 percent of this year's budget it would have immediately conveyed meaningful information to readers. And the editors at the NYT probably would have noticed if the piece had said that food stamps were 18 percent of the budget.

Anyhow, today the NYT obliged us with another example of meaningless budget numbers in reporting on the budgetary consequences of immigration reform. According to a new report from the Congressional Budget Office (CBO):

"The report estimates that in the first decade after the immigration bill is carried out, the net effect of adding millions of additional taxpayers would decrease the federal budget deficit by $197 billion. Over the next decade, the report found, the deficit reduction would be even greater — an estimated $700 billion, from 2024 to 2033."

Okay, are we in for a budget bonanza from immigration reform? After all, those are pretty big numbers.

Well in the first decade CBO projects that we will be spending about $50 trillion, so the savings from immigration reform amounts to about 0.4 percent of projected spending. That's about 1000 times as important as John McCain's Woodstock Museum, but only about 3 percent of what we are slated to spend on the military. The $700 billion over the following decade is a somewhat bigger deal, but it is still only 0.9 percent of projected spending.

But the real question here is do the NYT's editors really believe that they were conveying information to their readers by publishing these numbers? What percent of NYT readers have any clue as to the size of projected federal spending or the size of the economy over the years 2024-2033? I am willing to assert that it is much less than one percent, which means that the overwhelming majority of NYT readers had no idea what the numbers in this article meant.

What is the point of putting numbers in an article that don't mean anything to the people who read them? The vast majority of NYT readers do understand percentages. The NYT could have made this information meaningful to these readers simply by expressing them relative to the size of projected budget. That information is readily available from CBO as well as many other sources.

What possible reason can there be for not writing articles in a way that conveys information when it easy and costless to do so?