The GiveWell Blog

Job training: Which would you grant?

A couple questions for you. I think Cause 5 is going to come down to these questions.

1. Would you rather grant …

A. A program that helps severely unemployed/undermployed people, with barriers to employment including past convictions & drug abuse, get jobs paying $8-12/hr with no clear career path, such as security guard / nurse’s aid / administrative assistant, or

B. A program that helps already-employed people go from their $8-12/hr jobs to jobs starting around $30k/yr with a clear path to at least ~$40k?

2. Would you rather grant …

A. A program that takes a general-interest population and gets them into relatively white-collar-ish jobs (administrative assistant on the low end; computer support specialist on the high end), at great cost?

B. A program that takes people who are already interested in and capable of a particular, narrower, more “blue-collar” career (nurse’s aid; environmental resource technician; truck driver), spending far less to get them into these (equally well-paying) jobs because it’s primarily about getting them certified?

For (1), we need to know more about the connection between income and living standards in NYC … I personally feel like $8-12/hr does not count as “self-supporting,” especially when supporting a family is an issue, and so I’m leaning toward (B).

For (2), I feel that as long as there are still more people who want help than there are funds to help them, we should help the “low-hanging fruit” first: the people who just need a certification to get the job they want. So, that’s (B) as well.

What do you think?


  • michael vassar on October 2, 2007 at 3:11 am said:

    For 2, strong B.
    I’d go much further and say that I’d MUCH rather work in a blue-collar job than in a white collar job given similar pay, career prospects, and benefits. I’ve been a waiter and an actuary (among many other things). No comparison. Blue collar work is generally fairly fulfilling, as you actually see some good you are doing, while white collar work is pure hell, but with long commutes to office parks and repetitive stress injury instead of fire, which is come to think of it, a large improvement. OK, more like pure Tartarus. I bet Sisyphus has some bad repetitive stress injuries and a really serious case of alienation.

    By contrast, I’d be furious to be stuck in a $10/hr job because I lacked a certification to do the job that I wanted and was qualified for. Been there too actually, except replace $10/hr with $50/hr. Still temporary misery, but I was able to put my wife through Columbia rather than being stuck in place, which is a big difference.

    For 1 I lean fairly strongly towards A based on what happiness research says about the direct happiness penalty for unemployment, and, to a lesser extent, based on my anecdotal experiences, which tend to confirm it (part of this is that a regular job limits the degree to which a person can indulge in many drug/alcohol addictions). One person can’t support a family in NYC on $12/hr, but two people definitely can. This is not even that difficult in Astoria, Harlem, or Inwood, especially if one job comes with medical benefits, but even without them. One person can also work 2 $12/hr jobs, which is awful, but less awful than not being able to adequately feed, dress, and lodge your kids. Very many $8/hr jobs lead to $12/hr jobs and terribly importantly, all lead to unemployment insurance, fairly quickly.

  • I personally feel like $8-12/hr does not count as “self-supporting,”

    I don’t know too much about this but here is my uninformed opinionated opinion (please attack it if you disagree). $8-12 does not seem like a lot in New York City, but it goes plenty far in Wichita. To what extent are people entitled to live where they want?

    I am very privileged, but I move around a lot for my work (I am a graduate student in theoretical computer science). I was born in Green Bay, WI, but when I was nine-years old moved to Wichita, KS when my father got a job there. I went to college in Boston, then England; now I go to grad school in Berkeley. However, over the summer I was in Mountain View at an internship. Now I am in New Jersey for the semester. I would most like to live in Kansas, but it does not hold the opportunities I am seeking. Second to living in Kansas, I would like to live in the same place for more than a few months. I do not like to move and it takes me a long time to make friends.

    My point is that many move when their current location does not provide them with the opportunity to live the life they desire, but another location does. Why not support the people who can make $8-12 and simply expect that they will move somewhere where they can support their families?

  • It would be great if people could earn $8-12/hr living in Kansas, but one of the reasons the cost of living is lower in Kansas is that wages are lower. According to a Bureau of Labor Statistics report, in Wichita, average annual wages are around $30k/yr, and in New York, they’re about $60k/yr. So, those earning $8-12/hr would likely earn ~half that in Kansas.

  • Marie B on October 2, 2007 at 4:39 pm said:

    I’ve been spending considerable time on trying to fund “evidence-based” prison reentry programs with employment elements in and around Oakland. While it appears that many funders would side with B, I opt for 1A because if administered properly (and that’s a sticky wicket), it’s a proven method for breaking what they call “the intergenerational cycle”. When you factor in reduced costs in the prison warehouse industry, it would produce enough savings to fund both approaches adequately. But the underlying issue with any “employment training program” is that the legislated monies have a cut taken by the the Federal, State, County and local municipality before the remainder gets doled out to well-meaning nonprofits who do the actual work and are lucky to pay their people $8-12…

  • Elie,
    I do not find your logic altogether convincing. What you care about is not the average income (or even the better but still inadequate medium income), but how much people in professions like “security guard / nurse’s aid / administrative assistant” make. While they certainly do not get paid as much in Wichita as in NYC, you stats do not imply that it scales by factor of 2. Moreover, living is much cheaper in Wichita than NYC. Wichita is generally considered of the most affordable US cities (some fuzzy stat which compares income with cost of living). [American City Business Journals named Wichita second in their list of “10 most affordable places to live in the nation.–
    Do you really think that NYC is among the most affordable places for a “security guard / nurse’s aid / administrative assistant” to live? (I wish it were… I would be living there while going to school in Princeton now!) I do not think this is the weak point at which to attack my argument.
    So, another solution for GiveWell to research—relocation to Wichita KS!

  • Tim Wheaton on October 3, 2007 at 9:59 pm said:

    I think there are good reasons to consider answer choice B to question 1: that it’s better to help people currently employed in $8-$12 jobs get on the path to 30-40k/yr jobs. Since we’re trying to help both groups of people improve their employment situation–or “help them help themselves”–what’s necessary is to imrpove the amount of value they can add to an employer, since that’s what determines how much an employer is willing to pay them.

    So we’re now in a position where we want to improve the skill set of one of these two groups of people: those who already have $8-$12 jobs or those who don’t.

    It’s probably easier to cultivate the skillset of those with jobs already, since they’ve indicated some sort of skill just in virtue of having a job. Once you move these people up the skill ladder, it will then be easier for those without jobs to pick up the skilless ones left by those who move up. So if the objective is to help the worst off, it’s worth considering the possibility that helping the slightly better off is a more effective way.

  • michael vassar on October 4, 2007 at 12:08 am said:

    Interesting point Tim. More broadly, taking an economic perspective on the whole situation is sure to clarify the question being asked. However, I think that the best interpretation of this question is probably “is the difference in standard of living produced by 1A or 1B greater for the program’s beneficiaries”. Alternatively, “If you had an $8-$12 job and no better prospects were likely to become available, would you take a 50% chance of permanent unemployability for a 50% chance of doubling your income?” I think that the answer to this is a clear “no”, especially if the unemployability is coupled with drug addiction.

    Some data on unemployment causing unhappiness beyond the effects of reduced income.

  • Holden on October 5, 2007 at 3:11 am said:

    I think Michael’s approach (in the last comment) is the right one here. We need to know more about costs of living in NYC, so we can understand these numbers in terms of cost of living.

    I disagree with Tim’s analysis for a couple reasons. First, our goal is to help the unfortunate, not increase GDP/total wealth. The connection between the two is extremely questionable. I’m not saying there’s no connection, but you can’t tell me that increasing total wealth by $100 is reliably better than giving $20 to someone in need. More importantly, I should clarify that I’m trying to get at which you would prefer for the same cost. Based on our applicant, I think going from $10/hr to $20/hr actually costs more than going from welfare to $10/hr.

    Grant, I believe the higher wages in NYC are across the board, not just a function of the different distribution of professions. The same piece of property costs way more in NYC than in Wichita, because there’s more money floating around in NYC, period. Same for labor as for land. I will bet you that the same exact low-level job pays a LOT more in NYC.

  • michael vassar on October 5, 2007 at 7:20 am said:

    I’m sure Grant agrees that wages are higher across the board in NYC than in Wichita, but I think that he correctly doubts that the ratios are the same for high wage and low wage jobs.

    I’m very pro-immigration, but NYC has such a high level of immigration that it probably substantially depresses low wage incomes, while barriers to development and the fact of being an island radically increase New York housing prices.
    says that Wichita is 53% as expensive as NYC, but their minimum wage is the federal $5.85 while NYC’s is only $7.15 (and in 2009 both will be $7.25, as will all but California, Oregon, and Washington). It’s easy to confirm that there do in fact exist jobs in NYC that make less than $11.04/hr, the local equivalent of minimum wage in Wichita. In fact, there are minimum wage jobs in NYC. My father has one. Many people in NYC don’t even make minimum wage. After paying the cab rental fee (monopoly controlled with cab medallions, $120/day shift, $80/night shift I believe) and gas he often didn’t make even minimum wage as a cab driver, (night shift cab drivers also have one of the most dangerous jobs in the US

    Waiters, delivery people, etc routinely make less than minimum wage, especially at family restaurants. There are perpetual strikes near me by Saigon Grill delivery people over this.

  • Holden on October 7, 2007 at 9:25 pm said:

    I’m sure that in both Wichita and NYC, there are jobs that make less than minimum wage … seems fishy to assume minimum wage as the floor for Wichita, then point out that some jobs pay less in NYC, and conclude much about relative standards of living.

    NYC most likely has a wider distribution, period. More low-skill workers, more diversity of every kind. My point wasn’t that every single person in NYC has the same standard of living as the worst-off in Wichita. It was simply that a NYC admin assistant can’t necessarily move to Wichita and get the same pay plus lower prices. Again, I will bet that for the specific jobs under discussion, pay is higher in NYC.

    It doesn’t seem reasonable to solve NYC residents’ problems by demanding that they move elsewhere. This could work for some of them, but doesn’t seem like a good rule. For one thing, it isn’t clear that places like Wichita would exactly welcome an influx like this. And, staying put often has benefits (such as family, community and culture) that they won’t give up, and that are arguably part of a complete “standard of living.”

  • On the wage debate, Grant’s right that my 2x multiple was off. The median hourly wage for security guards in NYC is 20% more than in Wichita.

    I agree with Holden that asking people to move isn’t the best approach, but Grant’s question does make me wonder if it’s easier to help people become self-supporting in Wichita than New York City. If job training that will make someone self-supporting costs twice as much in NYC as Wichita, then it makes a lot of sense to focus there. Of course, we’d need to know more than just average wages and standard of living to make that decision. For example, while there are around 75,000 security guards in the NYC area, there are only 1,000 in Wichita. It wouldn’t be quite as easy to set up a program that trained and helped 200 people find jobs as security guards in Wichita as it would be in NYC.

  • michael vassar on October 9, 2007 at 1:30 am said:

    I think that it’s fair to assume that despite the existence of exceptions the vast majority of jobs in both NYC and Wichita are at wages over the local minimum wage.
    Really it makes more sense to ask particular people in the relevant cities, cite specifics a-la Elie and security guards, look at policies in chain businesses regarding what they pay in different locations, or look at local newspaper classified adds.

    would be a place to start, but newspapers are surely more useful in this respect.

Comments are closed.