@mrrich724 You seem to be speaking in a very general way. I asked if this data was compelling to you. Your response seems to be that it’s not, and that there is no data that possibly could be convincing because whatever it was, it could have been selected to prove a person’s point. I find this to be a very cynical point of view. And I agree with you that data can be cherry picked, and that statistics can lie. It’s just that I’m asking a specific question about specific numbers. If you can’t go to data at some point, how do you expect to be able to differentiate good policy from bad policy?
I suppose the broader question I might have asked is, do you think that the stimulus was a success? But I didn’t do that because I thought the conversation would quickly devolve into people arguing from their political side. The more interesting question, to me at this moment, is, does this piece of evidence support that statement, and do your own political leanings affect how you view it? The responses here fall into three basic categories:
1. Yes, absolutely.
2. No, or possibly not, because the data is presented in some misleading fashion, or because the data isn’t that relevant anyway.
3. The whole project of looking at data is pointless.
I absolutely respect the first two responses. I find the third response pretty weak, because it doesn’t respect my question, and because it’s just wrong. There is a great history of people looking at data with an open mind and having their opinions changed. I would argue that that, above all else, is what should inform our opinions. We need to be vigilant about misleading statistics, of course, and I respect the person who says the fact that it comes from a political headquarters makes them mistrust it… but as we’ve seen, the numbers are backed up by the bureau of labor statistics (thanks @laureth). I was partly asking this question so that people who don’t think the stimulus was a success would be able to tell me how they respond to this data. Keeping track of actual information, actual numbers, allows us to have some basis for our opinions besides just gut instinct, which doesn’t leave a lot of room for discussion. And this is something I want to discuss. I’m ready to have my opinions changed—you just need to do it with facts and arguments.
Are these numbers relevant? People seem to look at job loss as a key indicator of economic recession (example). If you were trying to come up with a policy at the start of this recession, one of the things you might have tried to do is figure out a way to stop job loss (at a reasonable price—yours is a very relevant objection @CyanoticWasp). If this is what the results looked like, you might determine your policy a success.
I’m curious what your political leanings are @mrrich724. And yours, @john65pennington. Again, if you have responses to this question that go beyond “all data are lies,” I’d love to hear them.
@Fenris and @liminal: I totally agree that there’s a complexity problem here—it’s very hard to know how to interpret these things. I think my question partly comes from a desire to set a clear standard. What counts as success? Is this an indicator that we’re on our way? Again, I can imagine people saying, “If we could just lower those job loss numbers, that would be a sign we’re moving in the right direction.” So I guess a further question for you, @Fenris, is, how do you determine good policy if the system is so complicated that it’s essentially senseless?
I hope that the inclusion of this second graph gives a sense of how large the job losses are relative to the entire US job market. For example, @njnyjobs, I think it’s clear that there are still plenty of jobs to lose, based on the other chart, and that the bars don’t shrink simply because all the jobs to lose have been lost. Mathematically speaking, that would account for only 1–3% of the difference in size, a negligible variation, in this case.