This post was written in connection with a larger post, Why I am voting third party, so I’m going to take a decidedly election-centric approach here. Sorry if you stumbled upon this looking for a more generic discussion. This is a long post that you may not agree with, so here is your trigger warning. Read at your own risk.
Hillary Clinton has aggressive plans to fight poverty. Her main way of doing this is by raising the minimum wage to $15/hr nationwide. So, my primary question is…will this actually affect poverty rates? My secondary question is…what unforeseen positive and negative consequences could result in? Let’s dive in.
I’m going to start with a little bit of general economics mixed with a bit of intuition. People have jobs because people buy things. People buy things because they have money. People have money because they have jobs. Pretty straightforward. If people buy more things, businesses need to produce more, so they create jobs. If people buy fewer things, businesses don’t need to produce as much, so they get rid of jobs. If people don’t have jobs, they have less money to spend, and will therefore buy fewer things. If people do have jobs, they have more money to spend, and will probably buy more things. If people make more money, they can buy more things. If people make less money, they can buy fewer things. If prices are lower, people can buy more things. If prices are higher, people can buy fewer things. To add some complexity to this, the people buying things doesn’t just include the American public. It also includes American businesses, the American government, people in other countries, businesses from other countries, and governments from other countries.
As you can see, all of these things are interrelated, and the extent to which each of them changes when a policy is enacted is what the study of economics is all about. Out of all of these, the democratic platform always seems to focus on wages. If we increase people’s wages, they can buy more things, and they’ll be better off. On the other side, Republicans and other opponents of minimum wage increases tend to only focus on job losses, saying that businesses will have to cut jobs in response to having to pay higher wages. As usual, both of these are incredibly narrow viewpoints. If your income doubles and prices also double, you are effectively no better off than you were before. If minimum wage doubles but you lose your job, you are no better off than before. Clinton, to her credit, does talk about prices on a few things. She wants to lower rent prices by building more housing. She wants to enact regulations to prevent price increases on pharmaceuticals. However, she says nothing of price increases at large or job losses that could come from many of her policies. This is somewhat ironic to me, because she actually argues that inflation (price increases) is causing the erosion of the value of the current minimum wage. Conservatives, for their part, do little to address the why or how much of job losses, ironically leaning on the Keynesian economic theories that they eschew in many other circumstances. So, that’s where I’m going to start my effort. I’ll build a case and I’ll work my way down from there.
Does increasing minimum wage lead to job losses?
For years, classical Keynesian economic models answered this question with a resounding “yes.” If there are people willing to work for less than the prevailing wage, and the prevailing wage can’t be decreased due to regulation, then there will be a shortage of jobs and unemployment will result. If the minimum wage increases, then more people will want to work more hours, but fewer businesses will be willing to hire people at that rate. I could show you a labor market supply and demand graph of this, but from teaching undergraduate economics, I know that no one likes those. Actually, why am I even discussing Keynes? No one wants to hear about him. Let’s just jump forward a few decades and dive into some new intuition.
The current conservative economic intuition says that if businesses suddenly have to pay their employees significantly more, business owners will either have to eat that extra cost through lower profits, raise prices to compensate, or lay off employees, all of which have negative consequences. I know that was my perspective on this for a long time. Of these three issues, it seems like laying off employees gets the most press.
On the other side, the current liberal intuition says that that employees who are paid more would be incentivized to work harder and stick around longer. This would lead to higher productivity, efficiency, and revenue as well. It would also decrease costs from employee turnover, offsetting higher wage costs so that employers wouldn’t need to fire anyone. Even if those things didn’t cover the increased wage costs, they argue inequality is rampant and business owners should share their profits with their employees.
So, which of these viewpoints that I have dramatically oversimplified is true? Well, in my research, I found that both sides significantly undermine the ingenuity and options of business owners, as well as missing many of the actual economic effects. John Schmitt performed a meta-analysis in 2013 for the Center for Economic and Policy Research that covered the results of dozens of studies on minimum wage increases over the years. From these, he identifies eleven different channels by which businesses can offset increased wages. These are:
- Reduction in hours worked
- Reductions in non-wage benefits
- Reductions in training
- Changes in employment composition
- Increased prices
- Seeking general efficiency and cost savings
- Cutting wages of higher paid employees
- Increased efficiency from employees being happier with higher wages
- Reduction in profits
- Increases in demand/consumer spending
- Savings from reductions in turnover
After reading through the explanations and various arguments for each and taking into account the areas where very little evidence was found, I’ve compressed them down. I cut reductions in non-wage benefits because near-minimum-wage employees rarely actually have non-wage benefits, and the studies showed this to have nearly no effect. I cut general efficiency improvements because businesses should be trying to do these anyway, so considering this as a catalyst is a bit ridiculous to me. I rolled reductions in training into the turnover topic.
Here’s what’s left:
- Changes in employment composition
- Reduction in hours worked
- Wage compression
- Increased demand/consumer spending
- Savings from reductions in turnover and having happier employees
- Reductions in profits
- Increased prices
Throughout this discussion, I’ll be using Walmart as an example. This is because they decided in 2014 to implement a two year plan to increase their minimum employee wage to $10. It appears that they used all of these channels to some extent, and so provides an interesting and timely case study.
Changes in employment composition
Conservatives often argue that the majority of the people making minimum wage are teens that need to build experience and are willing to work for lower rates because of this. I want to discuss this one right off the bat. If true, the likely outcome of increasing minimum wage will not be to address poverty, but simply to put more cash into the hands of teenagers. This topic is quite controversial, which is odd given that the IRS, Bureau of Labor Statistics (BLS), and the Census all have quite accurate data on how old workers are and how much money they make. The BLS statistics show that around 30% of people earning minimum wage are between the ages of 16 and 19. Another 25% are between 20 and 24 years old. This means that over half of all minimum wage earners are under 25 years of age. This doesn’t really support or undermine the republican argument.
Most studies do show that if anyone is losing jobs from an increase in minimum wage, it is likely the least experienced individuals, meaning these teens. However, these losses often are shown to come not from general job losses, but from changes in employment composition. The higher wage appears to attract more experienced and higher skilled employees, most of which are older. They get hired in place of teens due to higher levels of productivity. This says little about poverty on its own, though older individuals may be more likely to have dependents that could benefit from increased wages.
Reductions in hours worked
An increase in minimum wage is not technically an increase in the cost of having an employee, but is instead an increase in the cost of having an employee work for an hour. Due to this, employers could offset the higher hourly rate by simply cutting total hours. The extent to which this is viable depends on the particular business, of course. Walmart is reported to have cut the hours of many of its employees after raising wages. Individuals who earn minimum wage are largely concentrated into the leisure, hospitality, and retail industries, most of them in restaurants and retail outlets. While it may not be of great concern to Walmart, I have to think that decreasing hours worked would affect the operating hours and the total revenue amounts for many of these businesses.The evidence from various research shows that reducing work hours is still a big question mark, but I definitely think it would be considered by many businesses.
This could potentially be great for employees, if they were able to work fewer hours and still make the same amount of money. That doesn’t really directly address the issue of poverty, though, as the employee’s real income may not change at all. However, it could address it indirectly by allowing employees to save money on childcare or even by allowing employees to take on an additional job.
I can’t finish up this section without bringing up automation, which would definitely serve to lessen the number of employee hours required. I don’t have time to touch that one, but I definitely think it will become a more and more attractive option in the near future.
Increased demand/consumer spending
Individuals with lower levels of income tend to spend a bigger percentage of their money and save less than those with higher levels of income. Because of this, an increased minimum wage will tend to put more money into the hands of people who will likely spend it. Walmart’s CEO Greg Foran noted this as one of the reasons the company raised its minimum wage to $10, saying “Our wage hikes give workers more money to spend in our stores.” Since beneficiaries of the higher wages are likely to spend that money, this should offset some of the higher wage expenses for businesses with new revenue.
So, the big question here is to what extent increased consumer spending would be enough to offset the increased costs from higher wages. Most people argue that it is impossible for the costs to be completely made up by increased demand, and it’s an argument that is hard to refute. Take the Walmart example…say an employee is now making $20 more per day and the cost to Walmart is that full $20. The employee can choose to spend the entirety of that new income at Walmart. With that $20, Walmart covers the cost of acquiring the goods it just sold the employee as well as the overhead that enabled the company to sell the item and keeps what is left over as profit. Walmart reports a gross margin of around 25%, meaning that 75% of that $20 goes to the companies that produced the goods. It also currently reports a profit margin of around 3%, meaning that another $4.40 already goes to paying salaries, mortgages, rent, and the like. After all this, only 60 cents is left of the $20 revenue. So, even if an employee spends the full amount of extra income he or she receives at Walmart, the company will, at minimum, still have to make up 97% of that, or $19.40, from one or more of the other channels. Other industries with low wages such as retail clothing stores and restaurants have much higher margins, but the numbers will still never work out completely.
That said, the situation I described only represents a static view of a single point in time. To really see the full effects of increases in consumer spending, you would need to use a dynamic model that can account for all of the moving pieces over time. Such a model would incorporate the velocity of money and the resultant multiplier effect, which are two of my favorite obscure economic concepts. They basically say that an increase in demand/consumer spending that is introduced by giving a consumer an extra dollar actually causes a further increase in demand of more than a dollar due to a portion of the dollar being spent over and over again. I don’t have time to give an example, but the implications of this effect are one of the main reasons democrats often argue that increases in spending are a better way to stimulate the economy than tax breaks. Politifact actually gives a pretty fair treatment of the topic, if you’re interested in reading more about it. While it is not likely that the spending multiplier will be high enough to contribute much to the discussion, it’s always fun to throw out.
Wage compression means decreasing the wages of employees at higher pay levels to offset the increases in wages for employees at lower pay levels. People generally don’t like having their pay cut, and therefore the main way the evidence shows this happens is through decreasing the size of pay raises given to higher paid employees. While I do not feel this would significantly offset costs, it does offer a channel that would not have any significant negative outcomes save for upset employees. Walmart implemented this as part of its efforts to offset costs related to raising its minimum pay scale.
One of the primary arguments I’ve seen used by supporters of minimum wage increases is that employees will be happier working for the higher wage, will be less likely to changes jobs, and that higher quality employees will begin to apply. The logic goes that this will lead to significantly less time spent in employee hiring, onboarding, and training, and that higher quality employees or employees that stay around longer will have more experience and therefore higher productivity. These factors would lead to lower costs and higher revenues, thereby offsetting the higher wage expenses.
These arguments actually make a lot of sense, and I think they’re very applicable in professional settings. I myself have seen how turnover in my own professional workplaces can lead to decreased productivity. However, I feel that they break down a bit when talking about many of the industries that predominantly hire low wage workers. To achieve gains through efficiency improvements, they must lead to actual reduced costs or increased revenues that prevent employees from losing their jobs.
Having higher quality, more experienced employees can definitely increase productivity and lead to more positive customer experiences. However, I question the extent to which these will actually lead to increased revenues in low wage industries. Take food service, for instance. High quality service and efficiency can go a long way towards getting repeat customers buying drinks and dessert in the restaurant business, but I don’t think it does quite as much for fast food. In retail, having more knowledgable, courteous employees could very possibly lead to more computer or television sales at Best Buy. However, I don’t see this having as big of an impact at The Dollar Tree or Ross, where people generally know what they want and there aren’t as many up-sell opportunities. That said, Walmart has actually seen customer satisfaction skyrocket and same-store sales increase following its wage increase, so perhaps there is something to this.
Searching for job candidates and going through a selection process can definitely take time, but the extent to which that incurs actual monetary costs or decreases revenues in low wage industries is difficult to determine. As previously stated, I don’t feel training is a significant expense in most low wage industries, so I don’t think significant cost reduction will come from that. The main way I could see these come into play is if the time spent on hiring and training take the businesses owner away from other cost reduction or revenue generating tasks, but quantifying that is nearly impossible.
One way cost reduction could definitely come into play is if reduced hiring or training requirements or increased efficiency means some employees are no longer needed. Businesses may choose to reallocate HR functions to the owner or another manager and get rid of hiring or HR officers. If two new employees are able to do the work of three previous employees, the business may choose to get rid of the third position. I always approve of increased efficiency. However, the goal here was to find ways that businesses could avoid cutting jobs, and this seems like a great path towards cutting jobs.
The evidence shows that these effects do indeed occur, with my thesis advisor Barry Hirsch having performed one of the key studies on it. However, they appear to be somewhat weak effects
Reductions in profits
Many individuals who support minimum wage increases argue it’s a means of achieving more equality in society. They contrast rich CEOs with minimum wage employees and argue that taking profits out of the hands of the 1% and giving it back the employees is something that should be pursued. Thus, we have this channel, where instead of getting rid of employees, business owners will eat the costs of higher wages. While this is a valid way of avoiding firing employees, it does have its own set of consequences.
First off, many businesses that hire minimum wage workers are small and individually owned. I’ve watched Gordon Ramsey yell at enough small restaurant owners to know how tough the restaurant business can be on both the employees and the owners. There are plenty of rich small business owners out there, but every source I’ve found, and I’ve found many, shows that the average small business owner makes around $60k to $70k annually. While this is nothing to scoff at, it still only puts these individuals solidly into the middle class, and that says nothing of the huge amount of responsibility and hours many of them work. Granted, this only covers salary and does not speak to net worth or accumulation of assets, so the number is not completely helpful.
The second consequence of reduction in profits is that it may discourage future entrepreneurs from taking on the risk of starting a business. Decreasing profits means the returns to capital are lower, a fact that will likely make many individuals wary as this decreases the rewards for taking on risks. Small businesses drive the economy, and decreasing the rate of individuals founding them means long-term economic consequences.
The third consequence of reduction in profits applies specifically to publicly traded companies. While some minimum wage advocates may view decreasing profits as sticking it to the proverbial man, the problem is that said proverbial man is just as likely to be the advocate’s parents as anyone else. Sure, wealthy individuals own millions of shares of large retail companies, restaurant chains, and the like. However, millions of other shares of these companies’ stock also sit in the portfolios of teacher’s pension funds, company 401ks, and nonprofit 403bs. Companies disburse their profits to shareholders in the form of dividends, and fewer profits often means lower dividends. Share prices are connected to dividends. Even if dividends do not immediately decrease, the expectation that will often lowers share prices, which means lower retirement income for thousands of people across the income spectrum.
Walmart’s wage increase was implemented in two waves, jumping up to $9 at the beginning of 2015 and then to $10 at the beginning of 2016.The company’s profit margin declined from an average of 3.5% to an average of 3% over this time. While this may not seem like much, it represents $2.5 billion in extra costs given that Walmart has half a trillion dollars of revenue every year. The stock experienced a significant drop in its share value at the beginning of 2015 that continued through the end of the year, eventually putting it at its lowest point in the previous 3 years. While it recovered somewhat in 2016 and has not yet decreased its dividend, the full long-term effects of the changes on profits are yet to be seen.
This is the final channel I’m going to discuss, but it is probably the most important. If owners of a business can’t find enough additional revenue or cost savings using all of the previously mentioned channels and are not willing to face reduced profits, they can always instruct the business to increase prices. This can have significant effects on various aspects of the economy, many of which I have already covered in my discussions on tariffs and illegal immigration. However, the main effect I want to discuss here is the effect of increased prices on poverty.
The primary stated objective of Clinton’s proposed minimum wage increase is to reduce poverty. To determine if something will help reduce poverty, it’s important to know how the U.S. defines poverty. As stated by the University of Wisconsin-Madison’s Institute for Research in Poverty, “the U.S. Census Bureau determines poverty status by comparing pre-tax cash income against a threshold that is set at three times the cost of a minimum food diet in 1963, updated annually for inflation, and adjusted for family size, composition, and age of householder.” So, poverty in the U.S. is entirely based upon the price for a family to buy food. Food is generally sold in retail grocery stores and at restaurants. Together, these two industries contain over 50% of the individuals making minimum wage or below. So, if increasing minimum wage leads to businesses increasing prices, and these businesses primarily sell food, and the poverty rate is based off of food prices, the level of income required to escape poverty will increase.
I hope it’s obvious now that the extent to which prices increase following an implementation of a new minimum wage law has serious implications for how much of an actual impact that law will have on any meaningful metrics. Walmart would be a great example for this if the data was there. Their entire business model is built upon the concept of low prices, so seeing them increase prices significantly would be a smoking gun of sorts to this channel. However, in lieu of having that data available, I’ll turn back to academia.
Studies have generally shown small increases in prices following implementation of minimum wage increases, but the scope of these price increases and whether they is difficult to determine. A meta-analysis from 2004 states that most of the literature up to that point had found that most studies found that a 10% US minimum wage increase raises food prices by no more than 4% and overall prices by no more than 0.4%. A study from 2010 found “a positive and significant impact of the minimum wage on prices. The effect of the minimum wage on prices is, however, very protracted. A change in the minimum wage takes more than a year to fully pass through to retail prices.” One study published this year found that raising wages to $15 an hour for limited-service restaurant employees would lead to an estimated 4.3 percent increase in prices at those restaurants. A regrettably named University of Missouri study found prices of both McDonald’s burgers and Pizza Hut pizza increase with the minimum wage, and that these increases are quite large, amounting to roughly 50% of the increase in payroll due to the minimum-wage increase. In the less trustworthy sources category, one paper found no increases, but I never trust authors that only cite papers they helped write. A working paper from a Ph D student at the University of Chicago is the only one I was able to find that focused specifically on grocery store prices, and it found that a 10% minimum wage hike raises prices in these stores by about 0.7%.
One last study
Before sharing where I land on minimum wage increases now, I want to bring in one more study. It’s from the group that is looking into the effects of Seattle’s move to raise the city’s minimum wage to $15, which is the same rate Hillary Clinton is advocating for. The wage increase is to take place slowly over a number of years, so the effects seen thus far are only covering the period in which the minimum wage increased to $11.
In this preliminary study, the team found:
- The minimum wage itself is responsible for a $0.73/hour average increase for low-wage workers.
- The minimum wage appears to have slightly reduced the employment rate of low-wage workers by about one percentage point. It appears that the Minimum Wage Ordinance modestly held back Seattle’s employment of low-wage workers relative to the level we could have expected.
- Hours worked among low-wage Seattle workers have lagged behind regional trends, by roughly four hours per quarter (nineteen minutes per week), on average.
- Low-wage individuals working in Seattle when the ordinance passed transitioned to jobs outside Seattle at an elevated rate compared to historical patterns.
- Increased wages were offset by modest reductions in employment and hours,
thereby limiting the extent to which higher wages directly translated into higher
- Seattle’s low-wage workers who kept working were modestly better off as a result of the Minimum Wage Ordinance, having $13 more per week in earnings and working 15 minutes less per week.
- We do not find compelling evidence that the minimum wage has caused significant
increases in business failure rates. Moreover, if there has been any increase in business
closings caused by the Minimum Wage Ordinance, it has been more than offset by an
increase in business openings [due to a better overall economy].
Where does all this leave me?
I’m a moderate-leaning conservative, and like many with similar beliefs, I came into this post with the assumption that increasing minimum wage will lead to job losses. What I found was a much more complicated situation. Minimum wage increases may lead to job losses among teens in particular, but they most definitely seem to lead to employees working fewer hours, lower turnover rates, higher quality applicants, increased customer satisfaction, decreases in annual raises for higher wage employees, slightly higher consumer spending, and definite price increases.
Given the multiple negative effects that came about because of an effective wage increase of only $13 a week in Seattle, I have to wonder if it was really worth it. Having never worked for minimum wage, I don’t know how much an extra $50-75 a month means to these individuals. However, this only counts as a real increase in wages and will only address poverty if prices don’t go. The current U.S. poverty line for a family of two (think…single mother or sole-income couple) is $15,930. Remember, this is calculated off of food prices. In order to erase the full amount of increased wages ($13 * 52 = $676), food prices only have to go up by 4% (676 / 15930). That is an increase of only 8 cents on a $2 loaf of bread or 12 cents on a $2.50 gallon of gas. I’m not convinced that this won’t happen.
I am interested in and somewhat concerned about what effects will occur when the future increases happen as well. As to whether I support Clinton’s policy of a $15/hr minimum wage nationwide, I’ll defer to the lead author of the Seattle study, Jacob Vigdor. When asked by NPR about the effects they saw, he had this to say:
What I can tell you is that to think one minimum wage is going to have the same impact everywhere at all points in time, that’s not really consistent with what we’re observing so far. Higher minimum wages are thought of as a way to maybe allow some of the spoils and the profits of society to be distributed towards the lower-income workers. And spreading those profits and that wealth around, it’s a lot easier in a town like Seattle, where there is some wealth to spread. And it might not work so well in a place that is uniformly higher poverty, doesn’t have as many of these tech sector jobs or other types of high-income employment to make it all work. So that is one thing that I can tell you. We are going to be paying close attention. One thing that we have heard from employers is that the minimum wage is working just fine for them now, but that’s not necessarily going to hold the next time a recession comes along.
I don’t think there’s enough actual data on the effects of large minimum wage increases on real earnings, prices, or poverty to merit taking the risk of raising it at the national level. Most of the larger studies come from the 90s and early 2000s, and technology and society have advanced significantly since then. If Clinton committed to waiting until her second term to implement such a change, during which time additional data on Seattle, Chicago, and other places where the minimum wage has been significantly increased, I would consider it. However, as it stands, I find it difficult to support a candidate who is blindly advocating for such a sweeping policy to solve a problem that the policy has not been demonstrated to have any significant effect on.