Between 2009 and 2012, people in almost every demographic group delineated by the U.S. Census Bureau saw an increase in usage of government programs, according to a report released Thursday.

There was only one exception: people over the age of 65. The percentage of that group that used a means-tested government program -- that is, a program that is available to people depending on their economic status -- stayed the same over the four-year period. In every other case, the percentage of the group using government programs went up.

We can compare the new report to the Bureau's 2012 report, covering the period from 2004 to 2007, to see how the earthquake that was the recession affected the use of government programs. So, let's.

Overall participation in government programs crossed the 20 percent mark between 2009 and 2010. (The figures below actually indicate average monthly participation during the year.) In 2012, 21.3 percent of the population used some kind of program -- a list that includes Medicaid, food stamps, housing assistance, general assistance, Temporary Assistance for Needy Families or Supplemental Security Income.

Between 2004 and 2007, some groups saw a drop. But that didn't happen in the wake of the recession.

Medicaid and food stamps (a.k.a. the Supplemental Nutrition Assistance Program, or SNAP) were the most common programs of which people availed themselves. The change in Medicaid usage between the two four-year periods was small -- an uptick of 1.3 percent between 2007 and 2012. The change in usage of SNAP, by contrast, was dramatic, nearly doubling over the same time period.

It's easy to see why. If we look at the annual change during the two four-year periods among each demographic group relative to the overall percentage using the programs,* we see that the biggest changes were among the unemployed.

We can break that out for SNAP separately. The trend is similar.

By the end of 2012, the unemployment rate was still two percentage points higher than it had been at the start of 2004.

This overlaps with other factors. For example, single mothers and people who didn't graduate from high school see a larger usage of government assistance programs than other groups, consistently. Much of the variability, though, occurs within the population of the unemployed.

Again: After the recession hit, everyone used government programs more. But the increase in the number of unemployed people coupled with the increase in their usage of food stamps appears to have been central to the overall rise.

* In other words, we took the percentage of each demographic group -- high school graduates, for example -- and subtracted the overall rate of program usage, letting us see how high school graduates compared to the population on the whole.