You should ask Putin33 about that.

Oh the spaces didn't copy over. Let me try again.

My thought is to run an experiment to find the elasticity of labour supply for low income workers, focusing especially on the lower bound near minimum wage. My hypothesis is that labor supply will become increasingly inelastic as it approaches minimum wage and may actually come to such a point where it changes to a negative slope rather than a positive one – meaning that as you lower wage, people will work more.

The reasoning behind this is that people will want to consume quantity Q of goods. Some of this is achieved through traditional welfare means. However, since welfare only partially fulfills Q, people have to work for the rest. As their wage rises from minimum wage, I suspect workers will initially choose to work less as their wages rise, instead consuming leisure, as they value it more than their low hourly wage (so long as they have at least Q in consumption). This will go on until a certain point, not too far above minimum wage, where hours supplied begins to rise with wage again, as per the normal labour supply curve. So basically, the very lower bound of the supply curve I am hypothesizing will look like the letter C (sort of). This means that if managers demand a certain number of hours work from their employees, they could, for a very few levels hypothetically, offer two different wages and get the same result.

I just wanted to post this idea and see what criticism there was from fellow econ majors here: whether you think its absolutely stupid, if there’s been work on this already, etc. I just wanted to air the idea out to someone before I potentially go and make a fool of myself with my professor haha.

There. Much better.

Sounds like a pretty cool idea. Just be extremely rigorous about not designing your experiment in a way that encourages your desired conclusions -- think through what might cause your hypothesis to fail, make sure that your experiment would be able to pick up on those eventualities, etc.

I am studying towards a phd in econ - haven't had the time to read your post thoroughly, but a few things to point out:

1- if you are using empirical data, watch out for selection/truncation effects arising from the fact that you can't observe the reservation wages of the unemployed. Heckman has a seminal paper on this.

2- There is a very rich empirical literature on the national minimum wage - check out Card and Krueger on ideas.repec - it may give you some ideas to have a look. But overall it is

3- Try to find a textbook like Varian and read about income and substition effects - I am pretty sure there was a section on a backward bending labour supply curve where income effects overtake substititon effects. But I think it happened at high levels of wage, not low.

Yeah, my problem is how do you find hours of labor supplied in the job market? Surely that will affect things at the lower bound since people are so keen upon getting any job at all.

My idea is to use the frequency of call-outs and no-shows for a single nationwide retailer (say, Target or Walmart) to represent how frequently people are willing to forgo work, averaging this with how frequently said people add extra shifts. This will get me a single number of hours above or below their scheduled hours. Ideally, all the data would be for people who work near a full week (35-40 hours) since people who work less already have ample leisure time and would be more likely to pick up shifts. Having worked at Target, I'd ideally like to get the info from them.

So, the variables I would need to weed out a lot of other factors would be state (also to account for different minimum wage levels), age, gender, position, and how long they worked there (I spent a while thinking about this in lecture today haha). I'm aware there's the possibility that I could unintentionally make the experiment so that I find what I'm looking for, but frankly I have no idea how the data will turn out....or if I could even get that data.

Interesting and informative points, anlari. I think I've read some of what Krueger has written on it, but as I've said I'm not all that well versed. Where are you going for your PhD and what specifically are you doing it in? I'm at the LSE for a year abroad right now and am hoping I can get into their Masters program.

not sure if I can be of any more help since I am a DSGE macro guy, but the metrics involved in these kind of problems can go wrong very easily, so I would be extra careful. I am in Cambridge & my first paper will be on the importance of monetary sovereignty in shielding countries from sovereign default crises.

I'm not advanced enough for DSGE, but I've always been interested in it, since I want to ultimately end up in an institution like the IMF or Fed. I only have Intro to Econometrics under my belt, so the metrics will be quite the challenge. If you get your paper published, send me the link to it, as it sounds really interesting. I assume you studied the Eurozone for that?

There has been some research on this topic (such as http://www.iegindia.org/workpap/wp265.pdf) , but I think if you keep the topic and even aim more specific, you will certainly have something to add to the literature. Maybe focus in on teen employment and if it follows this curve. Or focus on some specific demographic.

I am graduating this weekend with an undergrad in economics (focusing on labor economics). I had to do a senior thesis as well, and I did mine with economics as well. Just giving you a fair warning: Get started on it early, because unless you have taken econometrics / have some Diff Eq. knowledge, many projects are going to take a long time and effort (e.g. help from professors) when it comes to creating models and analyzing the data.

I can send you some applied metrics lecture slides if you want, but I am not sure if it is a good way to spend your time when you probably have exams coming up. Is this for a dissertation?

You will get a taste of New Keynesian models when you do a masters - it is simple algebra, but a hell lot of it. At the moment I am extending the small open economy New Keynesian model to introduce sovereign default risk & financial frictions, then I will compare the impact on a stand alone country and a currency union member - so it is not directly on the Eurozone, but that is how I have been framing my research question.

If you are interested in this kind of stuff, I am basically trying to extend the paper 'Sovereign risk, Fiscal Policy and Macroeconomic Stability' in this page https://sites.google.com/site/giancarlocorsetti/main to the open economy.

I don't have to do this until the fall, but I thought its never too soon to brainstorm.

But yeah....I'm in the midst of revisions haha.

If you are interested in lab or field experiments I took an extensive course with levity & list and run controlled experiments frequently out of the office. Feel free to pm. I also have some sources that I can pull out to pick your brains.

I just took my Intermediate Micro final, and all of the examples we had on Utility functions as related to decisions concerning leisure time decisions as related to wage rate showed the opposite of your hypothesis. I don't know what the basis behind the problems being structured that way is, just thought it might be relevant. Also I'm interested in working on an economics study myself, and it is somewhat related to what your topic. Let me know how it comes along, I may want to cite you if you find something interesting, haha.

Your thesis is interesting and draws some parallels to agricultural economics. Once the price of farm outputs start to fall, as they are now for many types of produce, farmers don't cut back on production to increase the price of their goods. Instead, once prices bottom out, the only way they can make any money is to keep planting, which in turn increases supply and lowers the price even more.

That should read levitt and list. damn you autocorrect.

42

@goldfinger - interesting idea. I'm not studying econ so feel free to ignore my comments.

I used to volunteer for the CAB which means friends of friends still ask me for help with benefit entitlement. One such person asked me for help recently. Her English is poor and she had never claimed any benefit before. Her relationship status had changed and she was now living alone.

She was earning above minimum wage and wanted to explore tax credits, housing benefit and council tax benefit. Her employer had advised that she could have a few more hours as and when they became available. She worked on average 26hrs a week.

She wasn't entitled to tax credits as she needed to work at least 30 hrs per week. She was entitled to some housing benefit and help with council tax.

The point is she desperately wanted another 4 hrs a week and would have worked anywhere even at below min wage to get up to 30 hrs. But once that is achieved, her marginal rate of pay fell off dramatically. She would be paid at min wage for extra hours but benefit would be withdrawn.

In effect her decisions weren't based on gross wage but upon take home after benefit adjustment. Therefore first 26hrs at above min wage were crucial to her, the next four hours at below min wage were imperative, but the next 1 hr at below min wage would almost be not work it. It might equate to 15% of min wage.

The point is welfare affects the decisions more tHan the rate of pay. Any research has to include the cliff edges caused by welfare.

I once asked my professor about something similar, although not in context of a welfare state (minimum wage, some autonomous consumption guaranteed) so it wasn't about a consumer wanting to consume at least a consumption bundle Q, but about making enough to keep oneself alive. Surely one would try to work as much as possible if that was the only way to keep oneself alive? Well, it's actually difficult to measure via labour market since at superlow wages, the labourer would probably retreat from the labour market completely and revert to something like subsistence farming, begging, stealing, activities that bring income but are not based on offering one's labour on a market. Of course in your case, the minimum wage should eliminate most of this. Yet the problem is that the minimum wage is set at an arbitrary level and therefore acts as a cutoff since as aniari pointed out, you can't distinguish the reservation wages and therefore your sample will contain workers who would "gladly" work for less than minimum wage and their greater income per hour might put them on the backwards bend of their personal labour supply curve, so they will actually work less. So I think your hypothesis won't probably hold on the labour market as a whole, since the mass of workers eliminates the abberations on the lower and higher end. So while a personal labour supply curve might be double bended, resembling the letter S inverted by the y-axis (working more hours near the minimum wage, then working slightly less with increasing wage, then working more, then again working less), an aggregate curve will probably simply show a positive correlation between wage and labour supplied, irrespective of the subset of workers chosen.

I remember something from a long time ago about this where my microeconomics teacher talked about the curve being a c under certain circumstances, if you would like i will go over my past lectures and find what he was on about for you?

@ava - thanks for the offer, I may take you up on it. Unfortunately, though, I don't see how I could run a controlled experiment on this sort of thing.

@Puddle - normally for most wage levels it would be the opposite that the substitution effect overcomes the income effect. However, it has been found at high wage levels that the labor supply curve is backwards bending (which intuitively makes sense, but I'd be interested to see the r-squared values for those regressions). So while in the middle of the curve it behaves as you were taught, I'm interested in behavior at the ends of the curves.

@2ndWhiteLine - precisely. Only through collective action can the farmers coordinate enough to lower their planting and raise their prices. Even then, without institutions to provide some level of certainty, game theory says it would never work. Likewise, without unionization or the interference of some other institution (i.e. the government) low wage workers are stuck in that same game theoretic trap.

@maniac - its situations like that which led me to the hypothesis. Without welfare, she may have been happy to work for 26 hours a week. With welfare though, she is willing to work more hours at a lower hourly wage.

@ulytau - its precisely because of the arbitrary level of minimum wages and welfare generosity that this is so interesting. I mean, hypothetically we could go state by state with different benefit levels and different minimum wages and try to map out a fairly accurate equation to predict low wage labor supply. That's beyond the scope of my paper, but it could be further research for later. And I think its because of those people who would be willing to work for less that the labor supply curve would be "forward bending" at the lower end since they would work for more hours at minimum wage than other workers normally would.

I hadn't thought too much about aggregate effects, since I can't hope to capture the whole labor market in a study. Perhaps though the labor supply curve would be "thicker" at the lower end? With more variation than towards the middle.

@Socrates - not now, thank you. I'll be bookmarking this thread for August though, so I may be in touch then.