Who Works for Minimum Wage
Today, thanks to one of my professors from undergrad, I ran across this page.
We need to always be careful when dealing with statistics, as they are remarkably easy to twist.
Now, I don't think upworthy is being dishonest, strictly speaking. But, when they conclude "This is why raising the minimum wage would make a huge difference for tons of families, especially those in which women are the primary breadwinners or single moms." they go too far.
There are a few problems with that claim:
First, these pie charts are in percentage terms. We don't have enough data to prove that "tons" of families will be helped, because we don't know how many people, numerically, are being represented. That said, setting "tons" as the standard is setting a standard a bit low. Suppose, for example, than an "average" family weighs, say, 400 pounds total (probably on the light side for American families), helping 10 families counts as helping "tons" of families. Still, I think it's a bad idea to make policy on the basis that we can find 10 families that would benefit. Costs do matter, after all, one would think.
Second, these percentages are as a percentage of people that make the minimum wage. That is, this answers the question "If someone makes the minimum wage, what are they most likely to look like?" which is very different from "Which group is most likely to make the minimum wage?" And we get a totally different rank ordering for the two.
What do minimum wage workers look like? (Sex and age)
(1) Adult women
(2) Adult men
(3) Teen women
(4) Teen men
What do minimum wage workers look like? (education)
(1) Some college
(2) High school diploma, no college
(3) No high school diploma
(4) Bachelor's Degree or higher
Which group is most likely to make minimum wage? (Sex and age)
(1) Teen women (6.4% of teen women earn minimum wage)
(2) Teen men (3.8% of teen men earn minimum wage)
(3) Adult women (1.6% earn minimum wage)
(4) Adult men (1.0% earn minimum wage)
Which group is most likely to make minimum wage? (education)*
(1) Less than high school diploma (9.2% earn minimum wage)
(2) Some college (3.5%)
(3) High school diploma, no college (3.2%)
(4) Bachelors degree (0.6%)
[* - These numbers are skewed by the fact that I have the number of people that earn minimum wage from 16 years up, but only have total educational attainment for 25 years and up - so these numbers are probably off quite a bit - and would tend to overstate the "less than high school diploma" and "some college" numbers. If we assume that all the teens haven't finished high school, the "Less than high school diploma" number drops to 3.6%, so that suggests the order won't change, even if the numbers do if we had more complete data. Data I used came from here and here.]
But, even these numbers are somewhat misleading. What seems more relevant isn't what percentage of teens earn minimum wage. Instead, it's what percentage of teenage workers earn minimum wage. Well, we have that data, too. So, let's look at that.
What group of workers is most likely to make minimum wage? (sex and age)
(1) Teen women (23.8% of teen women that work earn minimum wage)
(2) Teen men (17.3%)
(3) Adult women (2.8%)
(4) Adult men (1.4%)
What group of workers is most likely to make minimum wage? (education)*
(1) Less than high school diploma (10.3% earn minimum wage)
(2) Some college (3.8%)
(3) High school diploma, no college (3.5%)
(4) Bachelors degree (0.6%)
[* - Same caveat as before, but not as severe this time. Assuming all teen workers have less than a high school diploma, that number drops to a around 8%. So, the order doesn't change.]
So, why do we get reversals between the first set and the others? Simple: because some groups are more numerous in the workforce than others. If adult men are a huge share of the population and teen men are a much smaller share, it shouldn't be shocking that a particular minimum wage worker is more likely to be an adult man than a teen man. To really get that this, let's look at which groups are most numerous among employees:
(1) Adult men (74.4 million employed)
(2) Adult women (65.6 million employed)
(4) Teen women (2.21 million employed)
(3) Teen men (2.15 million employed)
It's not interesting that more adults than teenagers make the minimum wage. There are about 40 adult employees for every teen employee in the work force. What is interesting is the reversal between adult men and adult women between this list and the next.
Here's another listing. Number of people working jobs paid hourly:
(1) Adult women (35.4 million)
(2) Adult men (34.6 million)
(3) Teen women (2.1 million)
(4) Teen men (1.9 million)
Wait. But, this matches the original ordering exactly. What does that mean?
It means the listing of "most common groups among those making minimum wage" can be fully explained by "most common groups working for hourly wages". Which, well, isn't that interesting.
It also means that we should broaden upworthy's claim, because, really, it's just a subclaim from the larger claim that "An increase in wages would make a huge difference for tons of families." Which is to say... what we all already know.
Here's something else that's interesting:
Group
Percent Min Wage
Percent Unemployed*
Teen Men
17.3
26.0
Teen Women
23.8
22.5
Adult Men
1.4
6.9
Adult Women
2.8
6.2
[* - Important definition: "unemployed" by the statistic's definition means "looking for work and doesn't have a job" - it is NOT connected with collecting unemployment benefits at all. That's a common misunderstanding. If someone used to count as "unemployed" and now doesn't, that means either (1) they stopped looking for work or (2) they have a job now. It doesn't mean their benefits ran out.]
So, while it looks like women are "discriminated against" when it comes to employers paying them the minimum wage, it looks more like they're discriminated in favor of when employers decide to hire them (versus leaving them looking for work). To me, this seems to line up well with a story suggesting that men tend to hold out for higher wages more than women do. On the top end of the spectrum, the result is that men get paid more, resulting in "gender gaps" in salaries and wage levels. On the bottom end, the gap shows up differently. Men are less likely to accept a minimum wage job - ending up out of work. While women are more likely to accept a minimum wage job. So, women are more likely to work for minimum wage, but less likely to be unemployed.
But, nothing that I've said so far touches the big claim from upworthy. Would "raising the minimum wage ... make a huge difference for tons of families, especially those in which women are the primary breadwinners or single moms"?
I don't think the answer is nearly as straightforward as upworthy suggests. Before I get at the theoretical reasons why, let's notice some interesting points:
(1) Teenagers are more likely to earn minimum wage and more likely to be unemployed than adults. Is it reasonable to explain this on a "holding out" basis, like we did for men v women? Well, no. If teens were holding out, they'd be less likely to make minimum wage than adults. But, if teens aren't holding out, why are they unemployed? A holding out argument doesn't work here. What does work is a productivity argument. Teens have less experience and education and are therefore less productive. As a result, firms don't want to hire them or pay them as much. This does well explaining both the high teen unemployment and the high teen minimum wage rates.
(2) To back up the productivity claim, it's also interesting to note that those without a high school education are far more likely to be paid minimum wage or end up unemployed than those with bachelor's degrees. Same argument: education improves (or at least signifies) productivity. Lack of education suggests low productivity - thus low wages or unemployment.
All this points toward an important fact: employers pay attention to your likely productivity level when deciding whether to hire you and how much to pay you.
To think rationally about the minimum wage, we need to set aside the rather stupid assumption that employers won't react to the minimum wage in some way that is counterproductive to the goal of improving the well-being of the workers. So, let's catalog possible reactions. There will obviously be effects on employers, so let's think about what those might be.
(1) The employer can passively take the hit - this means lower profits. For some reason, many people would rejoice at this. (Profits have a bad name, somehow). What are the real effects of this, though? They are twofold (at least). First, a decrease in profits decreases the incentive for firms to expand - leading to fewer jobs at all levels - including low wage jobs. Second, a decrease in profits diminishes the firm's ability to expand by diminishing a source of funds that can pay for expansion. So, both of these tend toward less investment in the business and fewer jobs. If it is severe enough, the lack of investment may actually lead to the decay and eventual closing of the business. None of these are "good" from the workers' perspective, I would guess. In the extreme case, this can happen very quickly. For example, if increasing the minimum wage increased McDonald's costs by 20%, they would no longer be profitable at all - and would have every reason to simply shut down unless they could adjust some other way, like...
(2) The employer can pass the cost on to the consumer. Their ability to do this, though, is limited. After all, as they try to charge more for the product, at least some of the customers switch away from buying that product to buying others. So, some of the cost will probably hit profits (as described above). But, if the employers pass some of the cost to the consumer, we have to remember: consumers are people, too - and perhaps even the same people that got the increased minimum wage. In as far as that's the case, the increased minimum wage is really just equivalent to someone handing money between their left and right hands and counting it as "increased income". They're no better off than before the "raise". In as far as the customers are different people than got paid minimum wage, this will be a transfer from the customers to the employees, making the employees better off, but customers worse off.
(3) A very likely outcome is a restriction in hiring. This comes from the productivity argument above. If an employer's willingness to hire or pay is limited by a worker's productivity then, in effect, the minimum wage is a "hurdle" that a worker's productivity has to get over before the worker can even possibly be hired. The higher the hurdle, the fewer people can jump over it. Result: less hiring. Making those affected by the minimum wage potentially worse off if they can no longer jump the hurdle.
(4) The employer can try to make up for the increased wage in some other way - perhaps by cutting fringe benefits or other costs. On the whole, this will make employees worse off, as they have fewer fringe benefits or lower quality materials and a lower quality work environment.
In a free market, it is true that an increase in wages makes workers better off - but that's because an increase in wages happens because of increased productivity. So, the reason workers are better of is that they're able to produce more "stuff", and their wages rise as a result. So, in the end they're better off. This is very different from forcing wages to rise as a matter of policy.
We need to always be careful when dealing with statistics, as they are remarkably easy to twist.
Now, I don't think upworthy is being dishonest, strictly speaking. But, when they conclude "This is why raising the minimum wage would make a huge difference for tons of families, especially those in which women are the primary breadwinners or single moms." they go too far.
There are a few problems with that claim:
First, these pie charts are in percentage terms. We don't have enough data to prove that "tons" of families will be helped, because we don't know how many people, numerically, are being represented. That said, setting "tons" as the standard is setting a standard a bit low. Suppose, for example, than an "average" family weighs, say, 400 pounds total (probably on the light side for American families), helping 10 families counts as helping "tons" of families. Still, I think it's a bad idea to make policy on the basis that we can find 10 families that would benefit. Costs do matter, after all, one would think.
Second, these percentages are as a percentage of people that make the minimum wage. That is, this answers the question "If someone makes the minimum wage, what are they most likely to look like?" which is very different from "Which group is most likely to make the minimum wage?" And we get a totally different rank ordering for the two.
What do minimum wage workers look like? (Sex and age)
(1) Adult women
(2) Adult men
(3) Teen women
(4) Teen men
What do minimum wage workers look like? (education)
(1) Some college
(2) High school diploma, no college
(3) No high school diploma
(4) Bachelor's Degree or higher
Which group is most likely to make minimum wage? (Sex and age)
(1) Teen women (6.4% of teen women earn minimum wage)
(2) Teen men (3.8% of teen men earn minimum wage)
(3) Adult women (1.6% earn minimum wage)
(4) Adult men (1.0% earn minimum wage)
Which group is most likely to make minimum wage? (education)*
(1) Less than high school diploma (9.2% earn minimum wage)
(2) Some college (3.5%)
(3) High school diploma, no college (3.2%)
(4) Bachelors degree (0.6%)
[* - These numbers are skewed by the fact that I have the number of people that earn minimum wage from 16 years up, but only have total educational attainment for 25 years and up - so these numbers are probably off quite a bit - and would tend to overstate the "less than high school diploma" and "some college" numbers. If we assume that all the teens haven't finished high school, the "Less than high school diploma" number drops to 3.6%, so that suggests the order won't change, even if the numbers do if we had more complete data. Data I used came from here and here.]
But, even these numbers are somewhat misleading. What seems more relevant isn't what percentage of teens earn minimum wage. Instead, it's what percentage of teenage workers earn minimum wage. Well, we have that data, too. So, let's look at that.
What group of workers is most likely to make minimum wage? (sex and age)
(1) Teen women (23.8% of teen women that work earn minimum wage)
(2) Teen men (17.3%)
(3) Adult women (2.8%)
(4) Adult men (1.4%)
What group of workers is most likely to make minimum wage? (education)*
(1) Less than high school diploma (10.3% earn minimum wage)
(2) Some college (3.8%)
(3) High school diploma, no college (3.5%)
(4) Bachelors degree (0.6%)
[* - Same caveat as before, but not as severe this time. Assuming all teen workers have less than a high school diploma, that number drops to a around 8%. So, the order doesn't change.]
So, why do we get reversals between the first set and the others? Simple: because some groups are more numerous in the workforce than others. If adult men are a huge share of the population and teen men are a much smaller share, it shouldn't be shocking that a particular minimum wage worker is more likely to be an adult man than a teen man. To really get that this, let's look at which groups are most numerous among employees:
(1) Adult men (74.4 million employed)
(2) Adult women (65.6 million employed)
(4) Teen women (2.21 million employed)
(3) Teen men (2.15 million employed)
It's not interesting that more adults than teenagers make the minimum wage. There are about 40 adult employees for every teen employee in the work force. What is interesting is the reversal between adult men and adult women between this list and the next.
Here's another listing. Number of people working jobs paid hourly:
(1) Adult women (35.4 million)
(2) Adult men (34.6 million)
(3) Teen women (2.1 million)
(4) Teen men (1.9 million)
Wait. But, this matches the original ordering exactly. What does that mean?
It means the listing of "most common groups among those making minimum wage" can be fully explained by "most common groups working for hourly wages". Which, well, isn't that interesting.
It also means that we should broaden upworthy's claim, because, really, it's just a subclaim from the larger claim that "An increase in wages would make a huge difference for tons of families." Which is to say... what we all already know.
Here's something else that's interesting:
Group
Percent Min Wage
Percent Unemployed*
Teen Men
17.3
26.0
Teen Women
23.8
22.5
Adult Men
1.4
6.9
Adult Women
2.8
6.2
[* - Important definition: "unemployed" by the statistic's definition means "looking for work and doesn't have a job" - it is NOT connected with collecting unemployment benefits at all. That's a common misunderstanding. If someone used to count as "unemployed" and now doesn't, that means either (1) they stopped looking for work or (2) they have a job now. It doesn't mean their benefits ran out.]
So, while it looks like women are "discriminated against" when it comes to employers paying them the minimum wage, it looks more like they're discriminated in favor of when employers decide to hire them (versus leaving them looking for work). To me, this seems to line up well with a story suggesting that men tend to hold out for higher wages more than women do. On the top end of the spectrum, the result is that men get paid more, resulting in "gender gaps" in salaries and wage levels. On the bottom end, the gap shows up differently. Men are less likely to accept a minimum wage job - ending up out of work. While women are more likely to accept a minimum wage job. So, women are more likely to work for minimum wage, but less likely to be unemployed.
But, nothing that I've said so far touches the big claim from upworthy. Would "raising the minimum wage ... make a huge difference for tons of families, especially those in which women are the primary breadwinners or single moms"?
I don't think the answer is nearly as straightforward as upworthy suggests. Before I get at the theoretical reasons why, let's notice some interesting points:
(1) Teenagers are more likely to earn minimum wage and more likely to be unemployed than adults. Is it reasonable to explain this on a "holding out" basis, like we did for men v women? Well, no. If teens were holding out, they'd be less likely to make minimum wage than adults. But, if teens aren't holding out, why are they unemployed? A holding out argument doesn't work here. What does work is a productivity argument. Teens have less experience and education and are therefore less productive. As a result, firms don't want to hire them or pay them as much. This does well explaining both the high teen unemployment and the high teen minimum wage rates.
(2) To back up the productivity claim, it's also interesting to note that those without a high school education are far more likely to be paid minimum wage or end up unemployed than those with bachelor's degrees. Same argument: education improves (or at least signifies) productivity. Lack of education suggests low productivity - thus low wages or unemployment.
All this points toward an important fact: employers pay attention to your likely productivity level when deciding whether to hire you and how much to pay you.
To think rationally about the minimum wage, we need to set aside the rather stupid assumption that employers won't react to the minimum wage in some way that is counterproductive to the goal of improving the well-being of the workers. So, let's catalog possible reactions. There will obviously be effects on employers, so let's think about what those might be.
(1) The employer can passively take the hit - this means lower profits. For some reason, many people would rejoice at this. (Profits have a bad name, somehow). What are the real effects of this, though? They are twofold (at least). First, a decrease in profits decreases the incentive for firms to expand - leading to fewer jobs at all levels - including low wage jobs. Second, a decrease in profits diminishes the firm's ability to expand by diminishing a source of funds that can pay for expansion. So, both of these tend toward less investment in the business and fewer jobs. If it is severe enough, the lack of investment may actually lead to the decay and eventual closing of the business. None of these are "good" from the workers' perspective, I would guess. In the extreme case, this can happen very quickly. For example, if increasing the minimum wage increased McDonald's costs by 20%, they would no longer be profitable at all - and would have every reason to simply shut down unless they could adjust some other way, like...
(2) The employer can pass the cost on to the consumer. Their ability to do this, though, is limited. After all, as they try to charge more for the product, at least some of the customers switch away from buying that product to buying others. So, some of the cost will probably hit profits (as described above). But, if the employers pass some of the cost to the consumer, we have to remember: consumers are people, too - and perhaps even the same people that got the increased minimum wage. In as far as that's the case, the increased minimum wage is really just equivalent to someone handing money between their left and right hands and counting it as "increased income". They're no better off than before the "raise". In as far as the customers are different people than got paid minimum wage, this will be a transfer from the customers to the employees, making the employees better off, but customers worse off.
(3) A very likely outcome is a restriction in hiring. This comes from the productivity argument above. If an employer's willingness to hire or pay is limited by a worker's productivity then, in effect, the minimum wage is a "hurdle" that a worker's productivity has to get over before the worker can even possibly be hired. The higher the hurdle, the fewer people can jump over it. Result: less hiring. Making those affected by the minimum wage potentially worse off if they can no longer jump the hurdle.
(4) The employer can try to make up for the increased wage in some other way - perhaps by cutting fringe benefits or other costs. On the whole, this will make employees worse off, as they have fewer fringe benefits or lower quality materials and a lower quality work environment.
In a free market, it is true that an increase in wages makes workers better off - but that's because an increase in wages happens because of increased productivity. So, the reason workers are better of is that they're able to produce more "stuff", and their wages rise as a result. So, in the end they're better off. This is very different from forcing wages to rise as a matter of policy.