As Justin Wolfers correctly stresses in the New York Times, your evaluation of the impact so far of the deep cuts to Unemployment Insurance (UI) benefits in North Carolina may have more to do with partisanship than with evidence. It’s a valid point, but it downplays a key argument regarding the cuts: that they would create undue economic hardship among the thousands of North Carolinians who lost benefits at a time when labor demand was, and remains, weak.

Wolfers focuses on job growth, which hasn’t come anywhere close to the breakout pace N.C. opponents of UI predicted. Their argument, as Wolfers describes it, was that a bunch more people seeking jobs would somehow create jobs (one hears the same thing in Paul Ryan’s recent poverty plan: requiring work is assumed to generate work). Yet job growth in NC was similar to that in surrounding states, which presumably face similar economic conditions.

Of course, that fact doesn’t help those who opposed the cuts on the basis of how much they would hurt the N.C. economy. The liberal claim — that as UI beneficiaries had less cash to plow into local N.C. economies, growth and jobs would suffer — doesn’t jump out of these data either.

Dean Baker argues that the conservative argument was more wrong than the liberal one, as the number of benefit losers is more concentrated in the state’s labor force than their more diffuse lost spending. Therefore, if the conservative argument was right, we should have expected faster job growth but not so much of a negative multiplier effect from reduced spending.

That’s reasonable, but so is Wolfers’ basic point that the data thus far do not offer anything like robust support for one side or the other. I’d add that his view is a useful reminder that it actually takes a while to assess such impacts — such state-level data are particularly sparse — and we should be wary of jumping on the first signs of evidence that we think makes our case.

But here’s the thing: To many of us who were watching this unfold, the danger of the N.C. cuts was much more micro than macro. It was about what the loss of benefits would do to the incomes of former UI recipients in a climate of weak labor demand.

The figure below shows that one thing we can be certain of, data-wise, is that a lot of unemployed people in N.C. lost UI benefits. The metric here is “recipiency rates,” the share of unemployed in the state who are receiving UI benefits. The left axis shows that N.C.’s rank among states (and territories) went from the middle to the bottom of the pack. The right axis shows the rate fell by almost half.

The data, provided to me by Mitchell Hirsch from the National Employment Law Project (full disclosure: I’m on their board) also show that for those who held on to UI benefits in N.C. in 2013, the average weekly benefit amount declined from $301 in 2013Q2 to $229 in 2014Q1.

Sticking with Wolfers well-advised humility about what little we know, I should point out that no one has tracked those who lost full or partial benefits to see the extent of hardship this has meant for them. But we do know, as per Wolfers’s analysis, that job creation in the state hasn’t accelerated, and that leads me to suspect, or at least worry, that we’ve made life a lot harder for a lot of people. If N.C.’s 2013Q1 recipiency rate had held steady at 42%, by the third quarter of last year about 70,000 more unemployed workers would have received benefits than actually did so.

I’m all about the macro and the multipliers, so I’m very sympathetic to the argument that there’s much still to be learned about that part of this debate. But let’s not lose sight of the fundamental role of this critical component of the safety net: to help working families get through periods where the quantity of jobs is simply insufficient for them to meet their basic needs. Given the lack of evidence suggesting that the N.C. labor market has suddenly and markedly improved, I’ll continue to worry about the impact of these UI cuts on the economic lives of those hurt by them.