Even more on the ineffectiveness of minimum wages as an anti-poverty measure

As was the case in Ontario, recent evidence from the US illustrates the pointlessness of using the minimum wage as a way to reduce poverty (h/t Craig Newmark):

Minimum Wages and Poverty: Will a $9.50 Federal Minimum Wage Really Help the Working Poor? Using data drawn from the March Current Population Survey, we find that state and federal minimum wage increases between 2003 and 2007 had no effect on state poverty rates. When we then simulate the effects of a proposed federal minimum wage increase from $7.25 to $9.50 per hour, we find that such an increase will be even more poorly targeted to the working poor than was the last federal increase from $5.15 to $7.25 per hour. Assuming no negative employment effects, only 11.3% of workers who will gain live in poor households, compared to 15.8% from the last increase. When we allow for negative employment effects, we find that the working poor face a disproportionate share of the job losses. Our results suggest that raising the federal minimum wage continues to be an inadequate way to help the working poor.

I understand that an internal study done for the Quebec government has found the same thing. I'll blog on it if and when it becomes public.

If you want an effective anti-poverty measure, give money to poor people. It really is that simple.

Comments

[Christopher Hylarides]

Stephen,

While I understand the efficiency of outright giving money to the poor to fight poverty, does it help the poor build actual wealth thus eliminating the need for such transfers or does it create a welfare trap?

I agree 100% on minimum wage. It destroys entry-level jobs. You don't see anybody pump gas for you anymore and it's rarer and rarer to see somebody bag your groceries. These were 2 very important starts for young people in high school. Hong Kong did very well for years and it had no minimum wages.

[Min]

"If you want an effective anti-poverty measure, give money to poor people. It really is that simple."

Beg pardon, but I doubt if it really is so simple. As Christopher Hylarides points out, there is the danger of a welfare trap, particularly when unemployment is projected to remain high for years. Being on the dole for an extended period of time is personally deleterious. In addition, there are costs to society as a whole. For a significant percentage of people to be on the dole for years drags down the whole economy. Skills deteriorate or, in the case of young people, are not developed. Attitudes deteriorate as well, leading to passivism and discouragement. And when people do get work, many will be underemployed, in work that does not make use of their abilities. In addition to the costs to society, workers learn to accept less and accomplish less.

Furthermore, with years of unemployment and the dole, people who have work become resentful of the chronically unemployed and blame them for not being productive members of society. This tends to make it even more difficult for the chronically unemployed to find work, as employers would prefer to hire people just out of school. The probability of producing a perpetual underclass is increased.

Better than extended unemployment is to put people to work. Don't just give them money. It is not that simple.

[Stephen Gordon]

Okay, perhaps not all *that* simple. The system of transfers has to be well-designed to avoid generating perverse incentive effects.

[Christopher Hylarides]

After reading Milton Friedman awhile back, I thought his negative income tax idea was the least of all evils (with it replacing minimum wage, welfare, food stamps, subsidized housing, etc) and scaling back the benefits in a way that if you started working, you at least ended up with significantly more money, minimizing the perverse incentive effects. While we have some tax credits that are roughly similar, we also keep on the books a lot of the other problem programs.

[tyronen]

Problems with this:

1) The US poverty line was devised in the 1960s and is based on food costs, which have fallen since then, and not on *housing* costs, which have skyrocketed since then. States like California probably have twice the poverty shown by official figures. (Canada doesn't even have an official poverty line, instead using LICO, which is more of an inequality measure.)

2) Minimum wage increases have a 'bumper' effect, pushing up wages even for people already earning more than the new minimum. The total effects can be much larger than many studies indicate.

3) Remember the moral dimension. It is *wrong* to pay people such low wages, subjecting them to hardship, even if they are 17-year-olds in a middle-class household. Not in a society where bankers are taking home multi-million-dollar bonuses, now at taxpayers' expense.

4) What Min said - the best social program is a job.

You have posted elsewhere that minimum wages under 45% of the average wage have no measurable effect on unemployment. In that case, the US minimum wage should be about $9.50 per hour, and I would continue to support that even if it had no effect on poverty.

[Darren]

tyronen: "It is *wrong* to pay people such low wages"

Can you explain to us how high wages have to go before it isn't 'wrong' anymore? Is there a formula, or at least a moral principle, involved?

[Christopher Hylarides]

I don't know what to say with tyronen's comments, but he seems to find it morally reprehensible that a teenager, who may only want to earn a bit of extra money, working for less than some amount that tyronen feels is just, is being exploited. What hardship exists there I don't know, but by that logic I was exploited by my parents by being forced at age 10 to wash dishes for about a half an hour each day for an allowance of $5/week (less than $1/hour!). Incidentally, if the allowance was negotiable to something I felt was worthwhile instead of imposed, I wouldn't have felt that way at the time. :-D This is to say nothing of farmer's children.

In our (Canadian) society, taxpayers aren't paying banker's bonuses (of which Canadian banks pay bonuses a fraction the size of US banks). Though the highest tax bracket in most of Canada is ~40%, so they're actually 'paying' us.

While pointing at widely successful companies and screaming that they can afford minimum wage increases (the vast majority of the most profitable ie oil companies, tech companies, and financial companies already pay way above minimum wage for most of their workers) it's easy to forget that most Canadians are employed in much smaller businesses such as retail, etc that have their costs squeezed by many sources, not the least of which is property and other taxes. Not all private businesses have endless streams of money and many will have to decide to let go of a worker to pay for the remaining ones that are getting the raises. Those businesses are subject to the same laws of supply and demand that the job seekers are. These companies are not evil, but are merely trying to survive. This is why nobody pumps your gas or bags your groceries anymore. Those jobs are now gone. Tyronen, you may want to actually see what it's like to run a business.

It is discrimination against those with low skills who cannot justify the higher wages. It prevents people from starting out and building skills. It's a good idea in theory, but in practice it only is good if you ignore the consequences you don't see in front of you.

[Andrew F]

I guess my only stumbling block is designing a negative income tax scheme that pays a decent (living) income at no earned income, without an unduly high marginal tax rate and without paying benefits until the national average income.

It seems to me that the least a person should receive is probably in the $10,000 - $15,000 range, even with a marginal rate of 50%, people would pay no tax at all until $20,000 - $30,000 income, which covers a large swath of the population. It seems that that would be a pretty expensive arrangement, and would require the rest of the population to pay a higher marginal rate (perhaps starting at 30% but closing in at 50% for high income individuals). I'm not convinced that savings from eliminating welfare, subsidized housing, multitudinous tax credits would actually amount to that much. It might reduce the cost of OAS somewhat, which might allow us to make the marginal tax rate less ridiculous for retired persons (as high as 70%). I haven't been able to find any analysis of what kind of system is fiscally possible. I think that such an analysis would provide some useful basis for debate. Maybe it's a job for the PBO!

[Christopher Hylarides]

Andrew,

In theory, it would replace the CPP as well. I don't know what level it should be at, but it should be enough for it to mean that if you choose not to work you're not going to lead much of a comfortable life.

Friedman obviously went really far with it (he was a libertarian after all) including removing all government support for health care, public schools, etc. In Canada this would require a lot of work if we ever decided to go that route. The theory is that people would best know what to do with the money themselves, whether it was buying health insurance, supplementing your retirement savings, getting childcare, or going to school.

I like the idea, but I don't expect to ever see it widely adopted. Most of the activists for current programs believe that the poor are poor simply because they can't help themselves or are taken advantage of by others. This is true for some (disabled, mentally ill), but many are caught in traps by refusing to move or disincentives by those programs.

[Andrew F]

I don't think eliminating the CPP is such a great idea, seeing as Canadians are catastrophically bad at saving for their own retirements, despite exceedingly generous incentives. I could see the case for reform, but I see a mandatory saving scheme as pretty essential.

I am not so hopeless as you about the idea. I could see centrists and conservatives being supportive of the idea, seeing as it would effectively reduce the size of the poverty industry (all the well-paid bureaucrats and less well-paid social workers, etc.), and reduce disincentives to work. I agree that the left is pretty hopeless, mostly due their general distrust of economic theory as a 'right-wing' discipline.

[David]

Interesting. I wonder if the Quebec study is available through access to information legislation.

[Stephen Gordon]

No, it's just not ready yet. I've been promised a copy of the final version when it comes out.

[tyronen]

Darren, Christopher: as Stephen has posted before, minimum wages lower than 45% of the average wage do NOT have an impact on unemployment. I am perfectly willing to pay more at McDonald's if the kid flipping burgers gets to actually sleep once in a while. I remember when I was a teenager and how so many of my peers slaved in junk jobs like that, frequently until 1 am or later, making $4 an hour. Were they exploited? Hell yes.

There are plenty of full-serve gas stations in Ontario, and I don't see what relevance children's household chores have to wage and labour market issues.

I could support the concept of a negative income tax, but I'd be astounded if the political right ever would - it would just be spun as giving welfare money to people who have jobs. We can't isolate from public opinion. People want to give the poor less money, not more. Expanded transfers aren't going to fly, no matter how much academic research backs them up. The minimum wage is popular, even in the conservative USA, because it is seen as rewarding work and effort, not sitting at home. And I'd rather push for a minimum wage that has a real shot of becoming law than a guaranteed annual income that has spent the past 30 years in fantasyland.

[RSJ]

"Can you explain to us how high wages have to go before it isn't 'wrong' anymore? Is there a formula, or at least a moral principle, involved?"

Keep raising the minimum wage up until there are inflationary pressures. Couple that with a Job Guarantee program and an end to welfare. If inflation ever goes below the target, then raise the minimum wage. If it goes above the target, then reduce fiscal spending and let the minimum wage deflate away in real terms.

[Christopher Hylarides]

Andrew,
If there was a negative income tax, it would be duplicating CPP anyways. The difference would be wealthier retirees wouldn't be getting income from the government anymore. Canadians are poor savers because we have things like CPP and OAS. In places without these programs, like China, they're complaining that they save too much.

RSJ,
A higher minimum wage wouldn't necessarily lead to higher inflation. Higher costs for McDonalds, etc may just leave people shopping there less or forcing them to rely more on automation.

Tyronen,
The vast majority of gas stations now are self serv and gone are all those jobs that teens often did after school during rush hour. A teenager isn't being exploited if he chooses to work those hours at that wage. Flipping burgers is boring work, but it's hardly exploitative. While there are exceptions where some kids are working to help support their families (one grocery store I worked at had a kid in this situation and the boss paid him under the table a higher wage), most of them are working to buy things that they want. You're right on people's support for the negative income tax. When they seriously talked about implementing it in the US a few decades ago, it was billed as exactly that: giving money to those with jobs. The minimum wage is popular, but I think it's misguided.

[Andrew F]

Job Guarantee? Is that welfare?

[Christopher Hylarides]

I wouldn't lob Job Guarantees with welfare per say, but they have an effect on incentives and most of them are created in unproductive work. They tend to create similar perverse effects as welfare and people stay in these jobs rather than risk trying to move up in case they fail. It still may be better than welfare, though various workfare plans had very mixed results.

The ultimate lesson is to always remember the law of unintended consequences.

[Patrick]

"giving money to those with jobs"

Funny that this would be perceived as 'wrong' as opposed to being an incentive to go get a job.

We have to keep in mind that poverty has a context. There's the obvious ones like addiction and mental health issues (both of which we could do much better at dealing with). I heard a terrible story recently on the CBC about a saw mill (I think it was a saw mill) in northern ON that was having labour problems. The company wanted wage concessions because business was in the tank, and the union refused. Lockout ensued (which seemed like an excuse to close an unprofitable mill), and the whole mess dragged on for months and eventually the company decided to shut the mill. And before anyone goes off union bashing, it seemed clear in this case that the economics of the mill just didn't work and it would have closed anyway. It was instant poverty for a few hundred people and their families. Where else is a guy with no education going to earn $60K a year in rural northern ON? They're hosed. What can you possibly do with a 45 or 50 year old guy, with no education, who has worked 30 years at the local mill? Families had already started to fall apart. The men end-up depressed, they turn to drink, smack their wives and kids around etc. It was really a horrible story, but one that seems to be all too common.

But the problem remain; what can be done? I don't see any answers, let alone easy ones.

[Andrew F]

With some negative income tax scheme, I think that a two parent family with say, two kids in the situation of one of those families affected by the mill closure, they could receive an after-tax income in the $20,000 - $35,000 range even with no employment income. Such a family should be able to get by with some lifestyle adjustments, and it would support the family while the parents retrained or searched for new work. It also occurs to me that this could reasonably replace EI (which is another wealth transfer scheme with somewhat perverse incentives), or allow it to be replaced with true employment insurance, with premiums based on employer and/or employee risk profiles.

As far as selling the right wing on it, I think reducing bureaucracy and simplifying the system in order to close loopholes and reduce gaming would be reasonably popular. I think it could be sold, especially if it started out fairly modest. I expect the most resistance from the left because it is not a paternalistic model and would also make many public sector 'poverty industry' jobs redundant.

[Christopher Hylarides]

Patrick,
While I don't know the full details of that case, but similar cases like that happen all the time. From the information you present, I would garner that the employees weren't properly explained to that they were choosing between unemployment or a reduction in pay and if they were they gambled and lost. "not giving an inch" is a common them among the more militant unions (especially CUPE). It's a story that's repeated as labour groups fail to learn that companies aren't going to be there forever and voters eventually get fed up at public employees. The members of the labour groups don't properly save money themselves, as they think the union will be there for them and get them what they expect. Some are learning to be more pragmatic. In the US, the auto workers unions are taking control of pension and healthcare costs themselves and leaving it to their members to top up the plans. That way they won't be left dry if companies go bankrupt or stop paying into the plans.

Incidentally, a negative income tax may have helped all these people/companies stay competitive.

Andrew,
That's just the way politics is. It's not even about ideology, but party hackery. Opposition parties tend to oppose anything the government does and how they spin it is how it gets entrenched in people's minds. Personally, I lean towards the free market and I thought the carbon tax was the fairest way to reduce greenhouse gas emissions, leaving special interest groups to have to compete fairly (cap and trade just benefits people who have already been inefficient in reducing their emissions). The conservatives spun it as a "tax on everybody!" even though the overall tax burden would have been less on those who were more efficient than the average. The conservatives wanted to win the election and they saw a way to make it a wedge issue.

[Andrew F]

True. Amusingly, Harper has mused about a carbon tax. I just roll my eyes. I don't care who's in power, as long as we get good policy. It's the bad policy I can't stand.

[Patrick]

Christopher: Yeah, I know it happens all the time. If I remember correctly, the union didn't believe the company when they said they needed wage/benefit concessions to stay in business. There not much room for an agreement when one or both parties think the other is trying to screw the other.

It's a bit OT, but my view is a little unorthodox in that I think labour and shareholders have much more in common they like to admit, and the real 'enemy' is so-called professional management, with its perverse incentives, rampant fraud, and rent extraction.

Unless we forbid corporation from forming associations, I think unions are probably a necessary counterbalance, short of all employees become owners (something non-employee shareholders really hate).

[Andrew F]

Why do non-employee shareholders hate employee share ownership? From what I hear, employee and particularly management ownership is a Good Thing, as it reduces agency costs. See, say, Westjet.

[Christopher Hylarides]

The balance comes from how much you dilute investor's who put up capital to give to people who are already being paid to do their job. If it's done too much, investors won't put in money in the first place and the company will have trouble raising capital to expand. With westjet, it may be worth it to keep labour peace in an industry that traditionally is bogged by labour problems.

Anyways, this will be my last input to his discussion. We're now way offtopic. :-)

[James]

I think the perception that minimum wage is a job killer on a grand scale is not well founded.

When rates went up across Canada, not just Ontario over the last few years (the one exception being BC); Unemployment fell in every region until last year. While the exact increases and their timing vary, we can say that most increases were between 20-30% above inflation, over the last 5 years, as catch-up was played for years of increases at or below inflation, and poverty reduction regained political fashion for a time.

No region suffered under this arrangement.

While, in theory one would expect to see pressure to raise productively in low-skill workplaces, ie. McDonald's this is hardly a bad thing, as society, a new-fangled automated cashier machine had to be designed, and built and serviced by someone making more than the minimum. Its primarly a shift to higher-skill labour.

If that means a few high school folks end up working slightly shorter shifs, while ultimately making just as much, as the min. wage is higher, that's hardly a social loss.

And it boosts productivity to boot.

I will comment on other aspects of this interesting discussion in a future post.

[Stephen Gordon]

Okay, but you should know that there's rather a lot of statistical evidence out there on the effects of minimum wage laws on employment in Canada; this is a recent, pertinent example.

But the point isn't about the effects of an increase in the minimum wage on employment. It's about the effect of the increase of an increase in the minimum wage on poverty.

[James]

I agree Steven there were other issues at play in the thread, which I want to address, I just didn't want to ignore that aspect of the issue, or to write a 20-page comment.

***

On to the issue of the minimum wage vs. poverty.

So to tackle this issue properly, it seems to me, we first have to have a working definition of poverty and some sort of sense of who comprises this group of 'poor' and why.

I must acknowledge at this juncture, I am aware of no scientific definition of poverty, nor any meaningful legal definition, at least in Canada.

As such I must sub my own in here, which will needless to say vary from any other person's given perception. I'm simply using my own definition in order to make my subsequent arguments clearer.

Poverty: The Failure to be able to obtain what would the social consensus would deem the most basic necessaries of life and lifestyle.

(crude detail, you should be able to afford at least a 1 bedroom apartment (modest, but safe, in good repair and hygiene, adequate nutrition, phone, a transit pass, some minimal clothing budget, access to adequate healthcare and education, and ideally a few very small luxuries, the Irish system I think notes you ought to be able to throw a modest dinner party once a month)

Obviously the exact things anyone puts in this basket may vary, and the exact costs are different in different jurisdictions.

However, using that as a base....

I will assert than for a childless person, poverty is earning below $20,000Cdn in rural/low cost areas, or $25,000Cdn in Cities). For each child add $5,000 required income.

So who are these poor people and how does minimum wage affect them?

Obviously many are children of the poor, so for now let's only consider those who are over 18, and under 65.

I would suggest that something like 1/4 of those in this income category are unemployed chronically and are managing on charity/social assitance.

3/4 are employed partially or fully, but typically earning the minimum wage or a few dollars p/h above.

NOTE: Using my earlier guess-work, minimum wage in Ontario, the highest in North America, will pay $10.25 p/h as at March 31, 2010. Which is roughly $20,500 per year, based on 40 paid hours per week.

We can clearly see that even those working full-time hours would very much struggle to get by, even with no dependents, especially in big cities/high cost areas.

Needless to say, those who get fewer paid hours, by choice or circumstance are in that much more difficulty.

NOTE: While labour laws vary in North American, paid meal breaks are often not required. As such, when most of us think of '40 hours' or 9-5, we assume paid meal breaks. For most hourly employees, if they are lucky and getting full shifts, 37.5 hours (30min paid meal break per shift) is far more likely. That means it becomes necessary to work 42.5 or possibly 45 hours per week to get back to the '40-hour' pay standard.

This inequity could be addressed through labour law, and paid meal breaks OR through assuming the minimum wage should be calculated assuming 37.5 per week of earning power.

******

FYI, in the Ontario example, at 37.5 paid hours, a min. wage worker will recover $19,900 before tax

*****

Now, back to the direct effect of minimum wage.

Let's for now assume that those in poverty who are either unemployed (whether by choice or circumstance)or have very limited hours, will not be affected materially by any minimum wage change.

So what we are talking about doing is affecting those adults, working age, who are either minimum wage earners or those in low-wage jobs whose compensation is set relative to the minimum wage.

Clearly, on balance the majority of these folks will be affected.

NOTE: Here is it is important to realize that if you factor out the chronically unemployed/underemployed, and any retiree-age poverty, that the most you will do is affect about about 60% of those who are poor directly.

***

What is the direct affect of a minimum wage hike on those types of people and jobs?

Of course there are so many job types, various work places etc. no example covers all affects.

But on balance, we can say the majority of minimum wage or near-minimum wage jobs are service-sector, typically retail in nature.

In a quick-serve/fast food model, labour represents about 40% of cost of goods sold. (the rest being reality, util,tax and product)

If the minimum wage rose by let's say $1 per hour, or a 10% increase in Ontario, this would almost certainly result in that exact increase across the board within that fast food outlet. Including for those making up $1.50 over minimum, in order to preserve the wage scale.

Let's note that this represents 4% of cost of goods sold, and can largely be absorbed unnoticed by the customer (the cost of Fries goes up .8c with a full pass-through)

For the workers, some of whom are poor, other middle-income family teens, it represents a big lift.

If you were one of those 18-64 year olds who needed their McJob, the extra income is $2,000 per year, enough in many cases to life you from under to over the poverty line.

While its true this increase would also benefit the middle income teens, and might encourage new productivity investment (say, self-serve pop machines); the teen takes an hours cut, from 20 to 18, and ends up with the same pay cheque as before. The productivity of the business per unit of labour is the same or higher, while the full-time older employee sees real wage growth.

I would argue this basic scenario of a minimum wage increase.

Moreover, it has the effect of encourage a few people from unemployed category to join the work force, as it becomes economically more attractive to do so.

***

One final note, the advantage of higher low-end wages to government are enormous. The hike offers the highest return on personal income tax for the government as most low-wage earners can't take tax deductions such as RRSP/401K, Mortgage (US), or RESP (Can.). They tend to pay the full marginal tax rate.

They also tend to spend almost 100% of every new dollar earned, and as such gov't sees soaring sales tax as well.

For the most part, min. wage increases do not materially affect corporate tax pay-out for the better or the worse, as many industries aren't affected, while those that are tend to compensation though pass-through price hikes, additional business (low-income owners w/more disposable income) and through productivity gains.

Gov't also tends to see a decrease in poverty program pay-outs when the minimum wage is on the rise (in the absence of other economic issues)

That win for government in turn allows it either expand programs or cut taxes or both.

[Andrew F]

So, a minimum wage increase flows mostly to middle-income teens, but it's still better than giving money to poor people? Consider that many middle-income teens will also be without jobs because their marginal product does not exceed the min wage. I think many do not fully appreciate the wage sensitivity of many service companies. My employer is very sensitive, and is investing heavily in processes and equipment to facilitate the reduction of low skill/low pay work. I'm not sure if that's a good thing or not. I lean toward not, if that means people who would like to work can't find a job, and that includes teens who just want some cash for an ipod.