The Minimum Wage Sausage
By Dan Willis
Jan 9 2016
Those of you that regularly listen to our podcast know that I have been advocating for an economic system that is more inclusive. In one of our shows, we highlighted a couple of companies that operate under what we will call the “democratic capitalist” model – a system through which the entire workforce, from rank-and-file employees to upper-level management, has a democratic say in the operations of the business.
Shareholders who invest in companies by purchasing shares in the secondary markets (the exchanges) are not adding any value. Zero. They're just becoming a stakeholder in the company's success (failure). But, importantly, their ownership of equity gives them a vote on board members and major strategic decisions, while the labor force of that company gets no say. The labor force must also assume the real-world changes that are affected by management and shareholders, and are therefore almost subjugated to the desires of shareholders. Economist Richard Wolff calls companies that provide their labor forces with a say in the action Worker Self-Directed Entities, or WSDEs. I think there is something to these businesses, and it’s what we’re doing here.
But this article isn’t about a democratic corporate America. This one is about the minimum wage debate (debacle?).
If you are a Republican (Libertarian, Conservative, “right wing-er,” et al.), you probably believe that hiking minimum wages has a seriously negative effect on employment. Your insistence is most likely rested on the notion that raising the minimum wage does more harm then good through passing costs on to consumers in the form of higher prices and lower employment. John Boehner, the former Speaker of the House (R – OH), has made his thoughts on minimum wage crystal clear: he once said he would rather commit suicide than vote for an increase. Whoa, man, take it easy.
Non-sequitur: why in the hell is Boehner always crying? I'm not sure whether to feel bad for him or not.
If you are a Democrat (Progressive, “Leftist,” et al.) you probably believe that hiking minimum wages has a somewhat negative effect on employment – but that it is the right thing to do. Policy makers that support higher wages for low-income earners are always bemoaning their lot in life: We are trying to help people at the expense of the economy, and it’s a super, super difficult job. It seems strange to me that someone advocating so fervently for a higher minimum wage would logically agree with their opposition.
So that brings us to the present framing of the minimum wage debate: from an economic perspective, it increases the price of labor – supply and demand theories tell us that when you increase the price for something, as a matter of rule, people demand less of it – but it might be the right thing to do. Or, it might mean economic disaster.
The argument is either, a) it will hurt businesses and consumers, which would result in reverberating negative effects, or b) it might hurt businesses and consumers, (resulting in reverberating negative effects) but it’s good for the employees at the lower end of the income spectrum. More often than not, you will here the first argument from Republicans and the second from Democrats. I say that to simply illustrate the fact that even Democrats kind of feel like they're robbing Peter to pay Paul, if you will, when they support higher minimum wage policies.
But I hear a lot of tangential conversation. Facebook turns into a minefield during an election year. One of the messages that seems to come out consistently is that economists have solved the debate: a higher minimum wage is a job killer. Is that true?
You Support a Minimum Wage Increase? You Socialist Idiot!
Let’s start with the case against raising the minimum wage. I found a wonderful little write-up from the Cato Institute, which outlines the four major reasons not to raise minimum wage:
Whew. That’s a lot of reasons. But, I want to highlight a couple of things about this wonderful little information sheet.
The Cato Institute linked 12 sources in this piece. 6 of the 12 sources came from Mark Wilson – who is this guy? He is the principal at Applied Economic Strategies, LLC, “where he provides economic and public policy analyses, and strategic advisory services to business, government, and judicial decision-makers to enable them to… make fully informed decisions.” Prior to starting Applied Economic Strategies, Mark was a Research Fellow at The Heritage Foundation. He has a B.A. in economics from Kent State University, “and continued with Masters Degree studies in economics at George Washington University from 1983 to 1987.” I think that means that Mark didn’t complete his Masters. And I am fairly confident that one cannot claim to be an "economist" without having earned one's Ph.D. Now, to be fair, he has more academic economic experience than me. But I do not believe that not being an economist should preclude anyone from attempting to understand their world and what makes it tick.
Half of Cato’s sources for this damning minimum wage piece are from a pseudo-economist. I am not sure why Cato did that; there is plenty of literature they could have used in support of their stance. Why they used Mark Wilson so heavily is head-scratching. I am sure Mark is a great guy but, strictly speaking, he is not an economist.
The last piece of interest from the Cato write-up is their reference of a 2006 review of more than 100 minimum wage studies by David Neumark and William Wascher. This is where I want to focus this article, because, as it turns out, the research done by Neumark and Wascher is in dispute.
In the arcane world of academic research, study-design is wholly important. If you design a study to examine the impacts of marijuana on the human body, you are going to need a control group that consists of people that don’t smoke any dope. That way, when you look at your experimental group, you have a clear picture to examine the effects of the drug against.
Now, in Neumark’s and Wascher’s review of 102 minimum wage studies, they identified a subset of studies that they considered credible – most of them are what are known as “fixed-effects models.” Fixed-effects models have one serious flaw: they do not have a comparable group to measure their results against. They lack a control group. In 2011, Sylvia Allegretto, Arindrajit Dube and Robert Reich used the Neumark/Wascher fixed-effects modeling framework, but included controls for state-specific labor market trends. What did they find? They found that, after controlling for state-specific trends, the standard negative effects on employment become “statistically indistinguishable from zero effects.”
Ok… So, What’s in the Sausage?
I love sausage. But, on the same token, I know that it probably is not 100% Grade-A pork (is that even a thing?). If you turned a piece of sausage into a pie-chart, it might look something like this:
If the economic research literature surrounding minimum wage policies were a piece of sausage, it might look something like this:
There is an underwhelming amount of comparative effects models in the literature. This is a problem. This muddies the waters a lot. Fixed-effects models use regression analysis, and these models rely heavily on assumptions. In a regression analysis, if one of your assumptions is even a little off, the entire thing is worthless. Not convinced? Look at how many regression-analysis-based economic models adequately captured the realities of the housing and debt debacle that led to the Great Recession. Spoiler alert: there are none. Alan Greenspan, Ben Bernanke, and almost every talking head on television turned out to be wrong, in tremendously embarrassing fashion.
So if fixed-effects models are questionable, how do we study this shit?
David Card and Alan Krueger wrote a paper in 1994 studying an increase in New Jersey’s state minimum wage from $4.25 to $5.05. They used a group of comparable restaurants in Pennsylvania, directly across the border from New Jersey. This was their control group. This painted a clear picture for their work: if any of the data from the New Jersey restaurants deviates from those of the Pennsylvania restaurants, then you have evidence of an effect from the minimum wage increase. It’s as close to a blind experiment as you can get in this field. Here’s my source.
They found zero evidence that the New Jersey minimum wage increases reduced employment when compared against the Pennsylvania (no wage hike) restaurants, sitting directly across the state border. This research was significant, and received a lot of attention – it was the first of its kind to study employment effects of minimum wage policies against a control group.
Of course, criticisms arose on the quality of the study’s phone survey data. But that is great; we want criticism of our ideas and studies. Card and Krueger returned in 2000 with more data confirming that the minimum wage increase in New Jersey had no statistically significant impacts on employment.
In the wake of Card and Kreuger’s initial 1994 study, some economists summarily dismissed the findings. They didn’t think that their work was based on sound economic theory. In reality, they didn’t find their work credible because they hadn’t logged a bagillion hours in Excel doing regressions. Economists love two things: mathematical models, and the traditional way of doing things. This research contained neither. But now, there is a much more productive conversation being had among economists to both highlight the importance of study-design, and figure out which design is best suited to measure the employment effects of minimum wage increases. There is a growing consensus among economists that when faced with the choice between a fixed-effects model and a comparison model, the latter is, unequivocally, the stronger study design.
Why This Matters
This matters because it is never mentioned. No one ever talks about study design. Also, it matters because this situation is nuanced. We cannot assume that raising the minimum wage will be detrimental, nor can we assume that it is the second coming of the Messiah. As we understand it right now, it is neither.
But what you can glean from this article is that the idea that raising minimum wages has demonstrably negative effects, immediately, on employment is, at worst, false and at best, in question. This should inform our responses to politicians and policy-makers when they make overreaching claims. I have this argument on a regular basis with other BrainBust members and friends. I am always confronted with the following phrase:
”It’s basic math, you fool! If you make widget “x” more expensive, people are going to want less of it! God damn, you’re a moron!”
I might be embellishing a tad.
Unfortunately for all of us, the complex world of economic cause and effect cannot be effectively boiled down into a meme. What a shame.
One Last Thought
The Pew Research Center put together a nice little graph on the buying power of the present minimum wage. Here it is:
The buying power of the current minimum wage is the same as it was in the 1960s and 70s. This completely re-frames the issue at hand. Raising the minimum wage is necessary to make sure those earning it can still survive on it. The ugly truth of this graph is that while minimum wage earners are taking home a higher nominal amount, the real buying power has failed to keep up with inflation.
One might be able to argue that the minimum wage should at least maintain its buying power.
But, then again, I’m just some guy saying stuff.
©Brainbust Media Group, LLC 2016