Great post. I've though a wage subsidy was better than a UBI for some time, although you don't hear much about it. I'm glad you discussed EITC because that is most commonly referred to as a wage subsidy. I also think it would be great to have a follow up with estimates of how much this would cost and how you would restructure other poverty programs.
Two questions: 1) what about someone who has one job with a wage above $16 and is thinking about getting a second job with a wage below $16 in order to supplement their income. Would they get the full wage subsidy on the second job. Would you need to do a true-up at the end of the year?
2) Let's say an employer is currently paying workers $14.25 / hour for a job. This wage subsidy goes into effect. Why wouldn't the employer immediately cut the pay to $7.25? Then with the subsidy, the worker is left unchanged, but the cost to the employer is nearly cut in half? You could say, maybe workers would complain to prevent this or the law would be such that it could prevent it. But maybe that happens initially, but what about in equilibrium?
Thanks for the comment and thoughtful questions! A follow-up with estimates of the total cost, as well as how this would interact with or restructure other poverty programs, is definitely on our list.
On your specific questions:
We are designing this policy at the employer–employee level. That means it doesn’t matter if a worker already has a primary job above the $16 threshold. For that job, they would receive $0 in wage subsidy. For a second job where the wage falls below $16, the subsidy would apply to the hours worked in that job. No year-end reconciliation is needed as workers would always see their true take-home pay directly in their paycheck.
Two factors prevent employers from simply cutting wages. First, because the subsidy covers only a share of the gap between employer wages and the target wage, lowering the employer-paid wage would also reduce the worker’s total take-home pay (employer wage + subsidy), so we can assume workers would recognize the cut as a real loss in wages, and not just a shift in share of compensation. Second, if a worker is worth $14.25/hour to their employer, rival firms will also likely value that worker at roughly $14.25/hour. If one employer slashes wages to $7.25, another firm could easily outbid them to purchase that workers labor. Competitive pressures should keep employer wages aligned with marginal productivity. In the short run, some experimentation may occur, but in equilibrium the employer wage (W) should continue to reflect productivity (MPL), even if the combined wage plus subsidy (W + S) may not equal MPL for qualifying jobs.
The response to the first question on workers with multiple jobs is reasonable. Makes it a lot more convenient for the government to manage. I think it can't hurt to comment on it when communicating to a wider audience. Basically what you said, "we're doing it this way for ease of implementation and simplicity for workers. A side effect could be that workers with relatively high wages in their primary jobs receive wage subsidies for second jobs.".
For the second question, I think you're right about the gap. I didn't take into account in my question that the worker who was paid $14.25 pre-subsidy now would now be paid $15.65 post-subsidy. They would be upset if the employer cut their wages from the post-subsidy value.
But the second part of your response was a little puzzling. If we are thinking about new workers or maybe new jobs that never existed before, then that morale issue from above doesn't hold. Let's suppose the labor is only worth $14.25 to the employer. Before the subsidy regime, they would have paid $14.25. Now they can say they will pay $7.25 with $7 subsidy. The worker wouldn't be lured away to another firm because their wage plus subsidy income is the same given that value of labor. The other firms who value labor at $14.25 would be doing the same thing. They would be offering labor at $7.25 with a $7 subsidy.
I think the response would be that "yeah, firms that employ low income workers are getting a subsidy." Oren Cass does that in the National Review piece you linked to. I think the question would be whether you are comfortable with workers getting that amount of a subsidy. The subsidy as a percent of wages (assuming wage subsidy is positive) is something like
K*(W_target/W_0 - 1) where W_target=16 and K=0.8 in your example. So when W_0 equals $7.25, it is on the order of 96%. For the subsidy to be 50% at an initial wage of $7.25 (keeping W_target=16), then you would need to set K to something like 41%. So about half of where you have it.
If the government announces a wage subsidy, it will certainly take effect at a specific time. Employers would simply cut wages the second that subsidy hits to match their previous wage paid to the employee. The employee wouldn’t recognize a difference at all. Last week’s paycheck was $14.25/hr, this week’s paycheck was $14.25/hr. You’re acting like every business owner in America wouldn’t immediately respond before the worker has a chance to receive or notice a difference.
This is an intriguing idea. Although one issue I have, wouldn't any raises that an employee gets between $7.25 and $16.00 be effectively taxed at a marginal rate of 80%, as the take home wage only increases by 20%? Granted, it still may be better on the marginal tax rate issue than other types of subsidies like the EITC.
1. Employee gets $14.25, $7.25 of which is from the employer and $7 from the government. Employee then spends $10 on "services" from the employer. Employer is better off, even with taxes. Employee is better off. The government has just been defrauded.
2. Same initial income, same split. Employee gets an offer from their employer: "hey, we really like the work you've been doing. Would you rather have a $4 raise or a dollar an hour under the table?" Employee prefers the money under the table.
My prior here is just that it's incredibly hard to fix all the fraud that wants to crop up at an 80% rate. 20-40%, sure, maybe some people game it but you can possibly structure taxes in such a way that it's fine. But at 80% I just don't think it's possible?
what do you think of the argument that those who earn minimum wage are likely renters. and that their landlords would effectively just raise rent to siphon any of this newfound wealth? are you familiar with henry george's argument about the land value tax?
This would massively distort the market by incentivizing hiring humans to do things that could be done more efficiently by machines. It punishes the most productive workers to reward the least productive, and thus disincentivizes being productive.
It’s a subsidy for low wages and employers who pay low wages. Employers would have little incentive to raise wages above the minimum wage and would be incentivized to create low paying jobs—which they already do! And have done for decades.
I literally just wrote that Dems need to talk more about JOBS. So I think this is great.
But I have a question.
Do both the median and lowest paid workers do better under full employment, when labor markets are tight?
In 2018 Annie Lowrey in the Atlantic argued that this isn’t true. "Across the country, there are more jobs available than there are workers looking for them, as the unemployment rate has dropped to a nearly two-decade low. Businesses are complaining of worker shortages, arguing they could do more and sell more and build more if they could just find the labor. Yet wages remain strikingly flat."
Also, if high demand drives up wages for low-wage workers, why do we see persistently low wages and worker shortages in the long-term care industry?
It seems to me that history shows bottom-half wages are persistently low even when labor markets are tight. Why is that?
Great and well-articulated post, but I'm afraid it's way open to fraud.
Putting aside the already mentioned cases (firm immediately goes from paying 14.25$/hr to 7.25$), I see another big one: firm "hires" Johnny for a non-existing job, giving him 7.25$/hr but requiring no work from him; in exchange, the firm requires Johnny to give them back 10$/hr; in the end, the firm earns 3.75$/hr and Johnny 4.25$/hr for doing nothing. How many thousands would jump on such free money?!?
Besides, you presume to give this wage subsidy even on multiple jobs for the same person; so Johnny could work 2 or 3 "imaginary jobs"... but still subsidized by taxpayers.
Remember: if something CAN be abused, it WILL be abused.
This is just more inflation on subsistence goods. Nothing in here about getting more of them, which is logical because paying people more money for the same work is going to result in zero change to productivity and therefore zero change in the availability of goods. Just like UBI and the COVID giveaways, the benefits will be inflated away in no time and we've just raised the bar for everyone.
Even if you taxed this amount from elsewhere in the economy, the net impact would be the same, since the concentrated demand would remain the same. Food, shelter and big screen TVs are the likely things to be purchased with this income boost. Rich people are spending their money on luxuries rather than subsistence goods. I mean other than the TV.
The only way I can see around this is to tax the very low income people that you are trying to match. If all the money came out of the pockets of the people already making $15 an hour, that would depress demand enough to make up for the giveaway on the other end. I don't think that fulfills the policy goal, though.
Figure out how to drive down the prices of subsistence goods like food, shelter, energy, clothing etc, and you'll do something meaningful about the plight of the poor. Government subsidies, carefully constructed, could do a lot about that. Giving them money will accomplish zilch except to spread the misery out even higher in the income scale.
Tax rent seeking. If speculating on fixed natural assets like location/land values was disincentivized with Land Value Taxes, then less of it would be held out of use. Landholders would be incentivized to make the most use of their land or sell it to someone who would. The result would be more housing and more businesses (and thus more jobs, products, and services).
If you could solve for the problem of tax avoidance, the idea I've mulled over for quite a while is to tax corporate wage inequality.
The tax rate would somehow be tied to the difference between the highest and lowest paid employees at a company. You could lower tax by paying your employees more. Or CEOs not having such high pay packages. That could leave more dollars for reinvestment or R&D. Its obviously not a fully thought out policy, but do feel like there's something to it.
Tax avoidance AKA "restructuring companies into wage bands" has, if anything, already become a prominent feature because, in part, because it's become increasingly awkward to treat your janitor that differently from your white-collar employees. So now firms contract out janitorial services.
I agree that in general, a wage subsidy is far preferable to most other programs that are designed to help the poor and near-poor. However, I think that you would get more positive results at lower costs if it were exclusively for married families with children and at least one full-time worker in the household.
Great post. I've though a wage subsidy was better than a UBI for some time, although you don't hear much about it. I'm glad you discussed EITC because that is most commonly referred to as a wage subsidy. I also think it would be great to have a follow up with estimates of how much this would cost and how you would restructure other poverty programs.
Two questions: 1) what about someone who has one job with a wage above $16 and is thinking about getting a second job with a wage below $16 in order to supplement their income. Would they get the full wage subsidy on the second job. Would you need to do a true-up at the end of the year?
2) Let's say an employer is currently paying workers $14.25 / hour for a job. This wage subsidy goes into effect. Why wouldn't the employer immediately cut the pay to $7.25? Then with the subsidy, the worker is left unchanged, but the cost to the employer is nearly cut in half? You could say, maybe workers would complain to prevent this or the law would be such that it could prevent it. But maybe that happens initially, but what about in equilibrium?
Thanks for the comment and thoughtful questions! A follow-up with estimates of the total cost, as well as how this would interact with or restructure other poverty programs, is definitely on our list.
On your specific questions:
We are designing this policy at the employer–employee level. That means it doesn’t matter if a worker already has a primary job above the $16 threshold. For that job, they would receive $0 in wage subsidy. For a second job where the wage falls below $16, the subsidy would apply to the hours worked in that job. No year-end reconciliation is needed as workers would always see their true take-home pay directly in their paycheck.
Two factors prevent employers from simply cutting wages. First, because the subsidy covers only a share of the gap between employer wages and the target wage, lowering the employer-paid wage would also reduce the worker’s total take-home pay (employer wage + subsidy), so we can assume workers would recognize the cut as a real loss in wages, and not just a shift in share of compensation. Second, if a worker is worth $14.25/hour to their employer, rival firms will also likely value that worker at roughly $14.25/hour. If one employer slashes wages to $7.25, another firm could easily outbid them to purchase that workers labor. Competitive pressures should keep employer wages aligned with marginal productivity. In the short run, some experimentation may occur, but in equilibrium the employer wage (W) should continue to reflect productivity (MPL), even if the combined wage plus subsidy (W + S) may not equal MPL for qualifying jobs.
Thanks for the quicky and detailed reply.
The response to the first question on workers with multiple jobs is reasonable. Makes it a lot more convenient for the government to manage. I think it can't hurt to comment on it when communicating to a wider audience. Basically what you said, "we're doing it this way for ease of implementation and simplicity for workers. A side effect could be that workers with relatively high wages in their primary jobs receive wage subsidies for second jobs.".
For the second question, I think you're right about the gap. I didn't take into account in my question that the worker who was paid $14.25 pre-subsidy now would now be paid $15.65 post-subsidy. They would be upset if the employer cut their wages from the post-subsidy value.
But the second part of your response was a little puzzling. If we are thinking about new workers or maybe new jobs that never existed before, then that morale issue from above doesn't hold. Let's suppose the labor is only worth $14.25 to the employer. Before the subsidy regime, they would have paid $14.25. Now they can say they will pay $7.25 with $7 subsidy. The worker wouldn't be lured away to another firm because their wage plus subsidy income is the same given that value of labor. The other firms who value labor at $14.25 would be doing the same thing. They would be offering labor at $7.25 with a $7 subsidy.
I think the response would be that "yeah, firms that employ low income workers are getting a subsidy." Oren Cass does that in the National Review piece you linked to. I think the question would be whether you are comfortable with workers getting that amount of a subsidy. The subsidy as a percent of wages (assuming wage subsidy is positive) is something like
K*(W_target/W_0 - 1) where W_target=16 and K=0.8 in your example. So when W_0 equals $7.25, it is on the order of 96%. For the subsidy to be 50% at an initial wage of $7.25 (keeping W_target=16), then you would need to set K to something like 41%. So about half of where you have it.
I don’t see how you addressed the problem at all.
If the government announces a wage subsidy, it will certainly take effect at a specific time. Employers would simply cut wages the second that subsidy hits to match their previous wage paid to the employee. The employee wouldn’t recognize a difference at all. Last week’s paycheck was $14.25/hr, this week’s paycheck was $14.25/hr. You’re acting like every business owner in America wouldn’t immediately respond before the worker has a chance to receive or notice a difference.
This is an intriguing idea. Although one issue I have, wouldn't any raises that an employee gets between $7.25 and $16.00 be effectively taxed at a marginal rate of 80%, as the take home wage only increases by 20%? Granted, it still may be better on the marginal tax rate issue than other types of subsidies like the EITC.
I don't get why you think you've fixed fraud:
1. Employee gets $14.25, $7.25 of which is from the employer and $7 from the government. Employee then spends $10 on "services" from the employer. Employer is better off, even with taxes. Employee is better off. The government has just been defrauded.
2. Same initial income, same split. Employee gets an offer from their employer: "hey, we really like the work you've been doing. Would you rather have a $4 raise or a dollar an hour under the table?" Employee prefers the money under the table.
My prior here is just that it's incredibly hard to fix all the fraud that wants to crop up at an 80% rate. 20-40%, sure, maybe some people game it but you can possibly structure taxes in such a way that it's fine. But at 80% I just don't think it's possible?
what do you think of the argument that those who earn minimum wage are likely renters. and that their landlords would effectively just raise rent to siphon any of this newfound wealth? are you familiar with henry george's argument about the land value tax?
This would massively distort the market by incentivizing hiring humans to do things that could be done more efficiently by machines. It punishes the most productive workers to reward the least productive, and thus disincentivizes being productive.
It’s a subsidy for low wages and employers who pay low wages. Employers would have little incentive to raise wages above the minimum wage and would be incentivized to create low paying jobs—which they already do! And have done for decades.
I literally just wrote that Dems need to talk more about JOBS. So I think this is great.
But I have a question.
Do both the median and lowest paid workers do better under full employment, when labor markets are tight?
In 2018 Annie Lowrey in the Atlantic argued that this isn’t true. "Across the country, there are more jobs available than there are workers looking for them, as the unemployment rate has dropped to a nearly two-decade low. Businesses are complaining of worker shortages, arguing they could do more and sell more and build more if they could just find the labor. Yet wages remain strikingly flat."
Also, if high demand drives up wages for low-wage workers, why do we see persistently low wages and worker shortages in the long-term care industry?
It seems to me that history shows bottom-half wages are persistently low even when labor markets are tight. Why is that?
Great and well-articulated post, but I'm afraid it's way open to fraud.
Putting aside the already mentioned cases (firm immediately goes from paying 14.25$/hr to 7.25$), I see another big one: firm "hires" Johnny for a non-existing job, giving him 7.25$/hr but requiring no work from him; in exchange, the firm requires Johnny to give them back 10$/hr; in the end, the firm earns 3.75$/hr and Johnny 4.25$/hr for doing nothing. How many thousands would jump on such free money?!?
Besides, you presume to give this wage subsidy even on multiple jobs for the same person; so Johnny could work 2 or 3 "imaginary jobs"... but still subsidized by taxpayers.
Remember: if something CAN be abused, it WILL be abused.
So you want the taxpayers to subsidize for-profit companies by paying part of their labor costs?
Why not raise the minimum wage instead of engaging in this corporate welfare?
This is just more inflation on subsistence goods. Nothing in here about getting more of them, which is logical because paying people more money for the same work is going to result in zero change to productivity and therefore zero change in the availability of goods. Just like UBI and the COVID giveaways, the benefits will be inflated away in no time and we've just raised the bar for everyone.
Even if you taxed this amount from elsewhere in the economy, the net impact would be the same, since the concentrated demand would remain the same. Food, shelter and big screen TVs are the likely things to be purchased with this income boost. Rich people are spending their money on luxuries rather than subsistence goods. I mean other than the TV.
The only way I can see around this is to tax the very low income people that you are trying to match. If all the money came out of the pockets of the people already making $15 an hour, that would depress demand enough to make up for the giveaway on the other end. I don't think that fulfills the policy goal, though.
Figure out how to drive down the prices of subsistence goods like food, shelter, energy, clothing etc, and you'll do something meaningful about the plight of the poor. Government subsidies, carefully constructed, could do a lot about that. Giving them money will accomplish zilch except to spread the misery out even higher in the income scale.
Tax rent seeking. If speculating on fixed natural assets like location/land values was disincentivized with Land Value Taxes, then less of it would be held out of use. Landholders would be incentivized to make the most use of their land or sell it to someone who would. The result would be more housing and more businesses (and thus more jobs, products, and services).
If you could solve for the problem of tax avoidance, the idea I've mulled over for quite a while is to tax corporate wage inequality.
The tax rate would somehow be tied to the difference between the highest and lowest paid employees at a company. You could lower tax by paying your employees more. Or CEOs not having such high pay packages. That could leave more dollars for reinvestment or R&D. Its obviously not a fully thought out policy, but do feel like there's something to it.
Tax what you want to incentivize.
Tax avoidance AKA "restructuring companies into wage bands" has, if anything, already become a prominent feature because, in part, because it's become increasingly awkward to treat your janitor that differently from your white-collar employees. So now firms contract out janitorial services.
Would it make sense to collect an employer tax to pay help, in part, pay for the program based on historical subsidies received by their workers?
I agree that in general, a wage subsidy is far preferable to most other programs that are designed to help the poor and near-poor. However, I think that you would get more positive results at lower costs if it were exclusively for married families with children and at least one full-time worker in the household.
I sketched out my version in this article:
https://frompovertytoprogress.substack.com/p/the-case-for-a-working-family-tax
What would be the estimated cost per year using the 21 million worker number at the beginning of your post?
Great question for ChatGPT…