In a Wall Street Journal column, Phil Gramm and Mike Solon ask whether the recent rapid expansion of the welfare state has altered the “old metrics” for assessing how the economy affects elections.

Specifically, they raise three pointed questions:

Do the tens of millions of Americans who use one or more of the 80 federal welfare programs “feel the same level of discontent about economic conditions as the rest of the voting population?”

“To what extent might these benefits not just foster dependency but also make the economy’s performance seem less of a deciding factor in voters’ choices?”

And, finally, “If you are concerned about your well-being and worried about a failed recovery — but getting new help from the government — do you vote for the candidate who promises more jobs or do you support the candidate who promises more government benefits?”

Excellent questions, all. But, before delving into them, let’s set the table with a few facts.
First, we should recognize that the War on Poverty is now a huge budget item. According to calculations by the Congressional Research Service and the Senate Budget Committee, taxpayers coughed up over $1 trillion in federal and state-provided benefits in 2011. These benefits flow to tens of millions of voters and cover the waterfront, offering low-income Americans everything from cash assistance to food, housing, and medical care, not to mention help with education, transportation, home-heating costs, and child care. Spending on these programs has soared more than 40 percent since 2007. That’s an unsustainable trajectory.

Other key facts, courtesy of Messrs. Gramm and Solon:

The percentage of the American labor force drawing disability benefits from the government has doubled since 1992, from 3 percent to 6 percent. They further note: “The number of workers qualifying for disability since the recession ended in 2009 has grown twice as fast as private employment.”

During the last four years, the Obama administration’s aggressive promotion of the food-stamp program has increased the number of recipients by 18.5 million.

Unemployment insurance that lasted no longer than 55 weeks in 1980 and 72 weeks in 1992 now can last 99 weeks. Some 40 percent of unemployed workers have been out of work for more than half a year.

The old political equation held that when the economy falters, incomes stagnate or decline, and unemployment remains stubbornly high, incumbents pay the ultimate political price. One way to assess whether this remains true is to examine how various voter groups look at the economy, which issues they prioritize, and how they view the two major presidential candidates.

The Battleground Polls conducted by the Tarrance Group on behalf of George Washington University and Politico make this level of detail readily available. The poll helpfully divides its sample of likely voters into, among other things, those who self-identify as either “low income” or “middle class.”

So, what do we know about these voters?

Those who self-identify as “low income” are more likely to be unemployed, frustrated over the state of the economy, and pessimistic over the general direction of our country than are those with higher incomes. Yet the Battleground Poll indicates they are more likely than those who identify as middle class to believe the country is heading in the right direction (42 percent vs. 35 percent).

Do welfare benefits insulate these voters from the sort of economic concerns that plague middle-class voters? Apparently so. Compared with their middle-class counterparts, far fewer low-income voters cite pocketbook issues as their number-one concern (53 percent vs.74 percent). Middle-class voters are, almost by definition, far more likely to pay taxes than low-income voters. Unsurprisingly, they are much more likely to list the economy and the level of spending and deficits as their most important concern (28 percent and 17 percent, respectively) than low-income Americans. Among the latter group, only 20 percent say the economy is most important, and a mere 7 percent worry about spending and deficits. Again, this is not surprising, considering that, for most low-income Americans, government benefits come with no strings attached, and at little or no cost in taxes.

In contrast, low-income Americans cite Medicare, Social Security, and education benefits as their number-one issue (29 percent in all) more than twice as frequently as do middle-class voters (only 13 percent).

If the receipt of welfare benefits affects voters’ views of the economy and alters the equation they use to judge candidates, one would expect them to give the president high marks for how he has handled the most stagnant and underperforming economy in over half a century. And, indeed, that is the case. By a margin of 51 percent to 37 percent low-income voters prefer Obama over Romney on this measure. They prefer Obama by an even more lopsided margin, 55 percent to 37 percent, on the issue of jobs. In contrast, Romney wins big among middle-class voters on these concerns (56 percent to 41 percent on handling the economy, and 54 percent to 43 percent on jobs).

“Based on the economy, Mr. Obama should lose on Nov. 6,” Gramm and Solon argue, “Yet, it seems implausible that tens of millions of Americans who have received additional government benefits during his presidency can be completely unaffected by that largesse. The election will test the relative power of private-sector aspirations and public-sector dependence.”

The Battleground Poll offers some real-time support for that proposition.

— Michael G. Franc is vice president of government studies at the Heritage