Friday, October 5, 2012

I don't understand this Paul Krugman post

Here.

Well, I understand and agree with behavior of the Phillip's Curve at low inflation. That's all fine.

But I don't understand his point at the end about average wages in the Depression. Are overtime wages "fake" wages or something? Who cares that when you take out overtime wages don't drop as much. How does that tell you that wages actually are stickier than the numbers imply.

Wage earners getting overtime sure view those as actual wages, and firms paying over time sure view those as actual labor costs.

[btw - often, but not always, when I say "I don't understand" something in the title, it actually means "I think I understand and I think you're wrong, but I'm not confident enough and/or too polite to come out and say that". I'm not really as humble as the title suggests.]

10 comments:

  1. Seems simple enough. Back in the days when I got paid overtime for working late, I've known bosses to say times are hard so we're cutting back on overtime. I've never heard anyone say here's the deal: we'll cut your regular salary, the good news is we'll let you keep your job. I know it happens, but it's a lot less common.

    ReplyDelete
  2. I think the point is that US wage falls in the early part of the depression are used as a counter-example to wages being sticky. But that it's actually contractual wages that are sticky and the US wage falls don't contradict this.

    ReplyDelete
  3. Krugman's point is that it is very difficult to renegotiate and reduce wage rates - that the rates are "sticky". There are statistics that show that realized average hourly wage rates during the depression fell. Krugman's point is that most of the fall in realized average hourly wages were a result of a reduction in over-time (which is paid at a multiple of the base rate and which does not require a renegotiation) rather than a renegotiation to a lower base rate for the wages. If you do not remove the overtime hours from the statistics you get a false impression of the stickiness of wages. (With Krugman, if you disagree with him, you should assume that you are wrong .)

    ReplyDelete
    Replies
    1. I get that sequence of statements.

      I guess it just sounds to me like he's saying "if you ignore the wages that do most of the falling and don't call those 'wages' then it turns out wages don't fall as much!"

      Delete
  4. OK, if the point is just that contractual wages are sticky that sounds fine.

    But I fail to see the macroeconomic significance of that sort of thing. Perhaps people don't make long term plans on the basis of overtime?

    ReplyDelete
  5. There is only so much overtime to cut. If you have to cut base wages to rebalance the economy and increase employment, for the most part, it isn't going to happen. Not sure how much overtime there was in southern Europe but there is massive unemployment. A flexible exchange rate helps a lot on that front, austerity won't.

    ReplyDelete
  6. The point is that neither overtime or regular wages declined that much. Your objection amounts to also saying that regular wages should be adjusted for the amount of employment since someone not working has had their wages cut to zero.

    Overtime wages did not change, overtime was just not offered.

    ReplyDelete
    Replies
    1. That makes sense.

      This is the only counter-argument that I think holds much weight.

      We have two sticky wage rates. Quantity adjustment occurs for the class of labor with the higher sticky wage rate first.

      It's not entirely clear why that would be, of course (why not cut workers making regular hourly and keeping workers who work regular plus some overtime?). But that's a sticky wage story.

      Delete
  7. I agree to some extent with Absalon, dcomerf and Andrew Bossie. Changes in overtime demonstrate that wages are sticky. But, overtime is in many cases a strategy that businesses use to compensate for that fact. The overall effect of the stickiness on the economy is still the combination of both normal hours and overtime, they can't be arbitrarily removed.

    This kind of thing can get quite silly. For example, if all you Keynesians are interested in making an argument for Keynesianism then why don't you eliminate overtime hours and all wages paid hourly. After all, salaries are more sticky than either normal wages or overtime wages. Looking at things the opposite way, Scott Sumner once told me that in New Keynesian theory (or at least his version of it) salarymen essentially don't exist, everyone is assumed to be a waged worker with some degree of control over the marginal hours they work.

    ReplyDelete
  8. I guess it just means you have to take that into account when evaluating feasibility of wage adjustment.
    In Italy, for example, the overtime bonus is ~10-20%, and the use of overtime in full-time dependent employment is minor, so you cannot expect great improvements to come from that side.

    ReplyDelete

All anonymous comments will be deleted. Consistent pseudonyms are fine.