An economist looking for a paper to write?

So, we\’ve a theory. That higher marginal tax rates will lead to people withdrawing their labour from the market.

We expect this to be greatest among women with children. For obvious reasons: the higher the marginal tax rate the more attractive looking after your own children will seem as opposed to working, suffering the tax bite, then paying someone else to do it.

So, to measure this, should we look at female labour force participation rates?

No doubt that (theoretically) high taxes could discourage effort but is this statement empirically relevant? Below is a chart of marginal tax rates (as estimated by the OECD) and the female employment to population ratio for the age range (25-54) for 2010. I have chosen that particular employment to population ratio because it matches the statement in the quote above (the chart looks similar if we look at a different age range or male participation rates).


Well, as he says, doesn\’t really seem to tell us very much.

So, here\’s the task for any budding economist who is looking for a paper to write. Instead of looking at labour force participation, which can be skewed by all sorts of things, cultural influences of course, part time working and so on, let\’s look at what it is that we really want to look at.

Which is we want to know, are these women substituting household production for market production in the face of these high marginal tax rates?

That is the original contention, of course. Not that women (or people in general, as above we just expect women to be more sensitive to this effect) work more or less in total in the face of taxes. But that they substitute away from the taxed activity to untaxed. That untaxed activity could be leisure, of course, or it could be household production.

Now, yes, we do have this information, for the EU at least it is here. How do people spend their time, in personal time, household work, market work and leisure? Split by country and sex (at least, there might be deeper with age group, family structure, not sure).

Marginal tax rates aren\’t that hard to find. So, a paper comparing *both* household and market working hours as against marginal tax rates.

Some of this work has been done in the LIS project (Smeeding is a name to conjure with I think) and I\’ve certainly seen one paper which shows that the average German woman is working more hours in total than the average USian woman. Despite the latter doing many more market working hours.

The result I would expect to see is that the higher the marginal tax rate the more subsitution there is away from market production to household production. And I\’d even expect to see, at times, longer total working hours for the \”same\” living standard at high marginal income tax rates. For market work is subject to the division and specialisation of labour, household production not so much.

To do this one needs to be able to play with Excel and graphs and charts and statistical tests on variance and SD and chi squared and so on. None of which I know how to do so, anyone looking for a paper to write?


15 thoughts on “An economist looking for a paper to write?”

  1. Um. Are you sure that chart doesn’t tell you much – or is it that it tells you something you’d rather not see? You’ve been happy enough to draw regression lines through scattered data before (no?). I think it is easy enough to by-eye this one: higher marginal tax leads to more female participation.

    Tim adds: I *never* draw regression lines. Don’t know how to.
    Next para of the source:

    “Do we see more or less effort in countries with high tax rates? Not obvious. In fact, in the sample I have selected there seems to be a positive correlation, not a negative one. ”

    The important part of which “in the sample I have selected”. Should have slightly more than 11 data points really, especially when 4 of our 11 are the Nordics.

  2. No one with an ounce of mathematical probity would stand by any line drawn through these data.

    I have to point out, as I’ve done before for many other figures, that the figure for Germany is complete bollocks, and this does not inspire much confidence in the other figures. We have, as a result of the pooling of couples income for tax (but not social security) purposes, an effectively very high state (both tax and social security) take on the marginal income. If the woman’s income is more “marginal” than the man’s the woman can easily be paying 65%. The rate might conceivably go over 100% if one considers the various baby freebies the state hands out (that are not all contingent on returning to work). This state of affairs is even enshrined in the tax system – in Steuerklasse V – whereby the party with lower gross income opts to pay a higher marginal (and indeed often absolute) tax rate than their partner. Sure, this all gets evened out at the end of the year and your choice of this system makes no difference to the money in the long run, but it does make it blindingly obvious just how much of your extra the state is taking.

    And because those social security payments are essentially a flat tax on the individual (not the couple), the whole thing amounts to massive disincentive to women to work – irrespective of kiddies.

  3. I am not an economist or had training in economics so that makes me eligible to be a politician, even a finance minister. However that aside how about this for a suggestion. The higher the marginal tax the more likely that that there will be higher percentage of females in employment as their extra income is needed to maintain a reasonable standard of living.

  4. Many working women are in the basic tax band(IMHO) so marginal rates will tell us little unless the data is specifically related to higher earning women.

  5. Looks horribly complicated to me:

    1) which tax rate? These things usually run off the top rate, but the lower rates directly affect far more people;

    2) include social security charges (NI etc);

    3) benefits withdrawal rates can be as important as tax, especially for women with young children;

    4) funny stuff, such as the German joint taxation that JamesV explains (#2);

    5) would you have to control for fertility rates? If the biggest effect is on women with children, then a lower proportion of women having babies would mean a smaller effect.

  6. The higher the marginal tax the more likely that that there will be higher percentage of females in employment as their extra income is needed to maintain a reasonable standard of living.

    You seem to be confusing marginal and absolute rates. The higher the absolute rate, the more you need to earn (which isn’t necessarily by working – benefits etc.)

    Marginal rate disincentive is proven at the benefits withdrawal rate: “I can’t afford to go to work” – especially child-care costs etc. It is assumed at higher levels and there is plenty of anecdata about, especially from business owners. What is missing from the current literature is the tricky analysis of reliable data so we determine how strong the effect is therefore what marginal rates can be tolerated.

  7. Thank you Surreptitious Evil. That proves I have not a clue when if comes to economics so I will definitely have to get a job as a politician or as a union official.

    Tim adds: I wouldn’t feel too bad about it. It’s exactly the mistake that R. Murphy made in the TUC’s submission to the Treasury a year or more back.

    Umm, on second thoughts, do feel bad about it.

  8. The marginal rate, and the fact that some taxes (NI in the UK, all the various social contributions in Germany) are taxed at an effective flat rate (for 95% of the working population) is decisive, when you have joint taxation. If Mr is working full time, earns a decent wage, his marginal rate can easily hit 30%, 40% or so. If Mrs then goes out to work as well, she pays 20% social security on her salary (roughly), and in effect the marginal tax rate – for we are really interested in the effect on net household income of Mrs not working. So working would bring in gross of, say, €2000 a month, but not working would mean an extra 30 to 50 hours a week, but around €800 less. That (with kids) could easily be less than the direct financial cost of working.

  9. > Tim adds: I *never* draw regression lines.

    I admit, I couldn’t find the post of yours I meant. Never mind.

    > Don’t know how to.

    Excel will do it for you, there is no need to think.

    > Next para of the source:

    “Do we see more or less effort in countries with high tax rates? Not obvious. In fact, in the sample I have selected there seems to be a positive correlation, not a negative one. ”

    Yes well, that was the point I was making: the data provide the answer, but it isn’t the one you or the person you’re quoting want. Had the correlation been the “expected” one, I rather get the impression that the regression line would have been drawn by your quotee.

    > … Should have slightly more than 11 data points really…

    Perhaps. Since you’re not lazy and have access to the same data and you care, then, errm…

    Tim adds: “Excel will do it for you, there is no need to think.”

    William, please do try and understand this. I do not know how to use Excel. No, not just being cute on the internet. I do understand what it is as a tool, I do understand the uses to which people put it. I even understand the progress from Visicalc to Lotus to Excel. But I *do not know how to use it*. I cannot enter a row of figures in. I cannot add an equation. I cannot do a calculation in Excel.

    It’s a bit like a piano to me. I appreciate their exiestence, love some of the results of their existence but this does not mean that I can play piano. Nor that I am willing to expend the effort to learn how to.

  10. There is definitely a significant segment of the electorate in the U.S. that would like to extend taxation of income to (mostly) women who stay home to keep house and raise young uns. They’d call it “imputed income.” It was a pet cause of Hillary’s back in the days when she was whomping up a health-care program somewhat secretly (and was at least temporarily shelved when the health-care initiative came under fire.

    It hasn’t gone away, though–nor would the idea be confined to the U.S. Politically, it’s attractive to working women, men whose wives work, and to most of the entire universe of benefits recipients. I’d guess the working women view it as a redress for sacrifices and the others as simply a source of taxes to lower their own (and/or strengthen the benefits pot from which they draw).

    It’s a mistake to blame politicians for these. All they do is frame the larcenous ideas of your neighbors and compatriots in reasonably acceptable fashion for publication and discussion.

    And I’ve got no idea how to make things better.

  11. I think Tim has mixed up his marginals. Most of us think of the 40% upper tax rate – or the new 50% rate – as the top marginal rate of income tax in the UK. (We really ought to count employers’ and employees’ National Insurance too.)

    Tim must be referring to the effective marginal rate for the poor coming off benefits and into work. In some circumstances these marginal rates allegedly approach 90% – i.e. for every £1 extra earned, they lose £0.90 in benefits.

    A worst-case scenario might be a single mother with three children living in an expensive rented house in London. While unemployed she receives a lot of housing benefit and council tax benefit; she receives free prescriptions, free dental care, and free glasses if needed; and the children receive free school meals and discounted leisure centre entry.

  12. Tim: The Nordics tend to do things like state subsidised cheap/free childcare, no? That must have an effect – in the UK, putting a couple of under 5s in full-time childcare costs about the entire net income of a woman with a reasonable full-time job.

    An academic friend of mine calculated that for a year or two she was actually worse off working (paying for decent childcare for 2 kids, commuting to work, buying appropriate wardrobe,…) than staying home with the kids. Husband works too, so state benefits aren’t an issue.

    She chose to work in order to stay in the game, so she still had a career when the kids were school-age.

Leave a Reply

Your email address will not be published. Required fields are marked *