Why do New Jobs Pay More?

An attempt to explain a bizarre economic reality

I believe that all of my readers found me through Freddie deBoer’s Substack. Imagine that, tomorrow, he announces his intention to quit blogging. In this announcement, he reveals that his blog will continue under a new author, and she will address similar topics like education, media, and politics. He then links to a couple of pieces this author has written, though the content doesn’t feel like anything deBoer has written. He also notes that the author won’t write much for the first six months, and he can’t guarantee that she will ever succeed as a blogger. Now, imagine the general reaction from his audience went something like this: “sweet, I'm going to pay extra for the blog now!” That’d be weird, right? Why would people pay more for someone who’s never done the job? Yet, that’s what damn near every company seems to be doing.

You’ve probably read a few stories about the Great Resignation. For the most part, we should call it the Great Hiring since the vast majority of quits involve someone leaving their current role to start a new one. Some of this stems from the re-opening of certain industries, like hospitality, while much of it has resulted from a general increase in economic activity.  However, even before the Great Quit And Then Get Hired At A Similar Position, everyone knew that changing jobs pays more than receiving a raise at your current one. At one of my previous jobs, the CEO admitted this fact when challenged about the company’s lower salaries. 

Isn’t that kinda strange? Sure, I took a labor economics class as an undergrad, and the professors stressed the differences between the labor market and the proverbial perfectly competitive market from econ 101. For one, the labor involves an uneven power balance. The worker needs the job more than the company needs the worker. Labor also involves a lot of imperfect information. It’s difficult for a firm to estimate the productivity of a potential worker, and it’s difficult for a worker to gauge the quality of various employers. There’s also large transaction costs, as anyone in their fourth round of interviews can testify. We also don’t see the level of homogeneity we’d expect from a competitive market. If one grocery store prices the apples too high, I can buy pretty similar ones from the store across the street. However, I can’t easily quit my job for one with a similar company culture, boss, co-workers, and location.

Still, there’s gotta be some logic, right? Of course, I don’t expect the econ 101 outcomes where wage=marginal product of labor. CEO pay has skyrocketed in recent decades, and I don’t imagine they’ve become that much more productive. One the other end of the spectrum, the poorest workers probably receive far less than their marginal product, which is why minimum wage increases don’t tend to cause large increases in unemployment. Is there no logic at all? Maybe the modern economy is too complex for anyone to manage, so wages reflect company politics, corruption, and other random nonsense rather than productivity. That position seems a bit too extreme, though I think there’s more bullshit in the private sector than most people appreciate. Maybe I’m naïve, but I’m not ready to dismiss the productivity-pay relationship entirely. 

If we don’t, however, I’m kind of at a loss. At your current role, you know the systems, the software, the points of contact, etc. At a new role, it can take a worker months to produce anything of value. Even then, you’ll have less experience at these new systems than they did at the old ones. That’s just how time works. There’s also the risk that a new hire doesn’t pan out. Maybe they exaggerated their skills, or they’re just not a good fit for the company culture. In short, it seems like a worker’s expected value in their current role exceeds that of a new role. Yet, workers consistently receive more compensation at that new role. 

Below, I will discuss a potential list of causes for the job-switching wage gain and why none of them feel satisfying. These explanations will only concern middle to upper-middle income white collar roles, since I haven’t thought much about anyone else.

Matching

Workers may be more productive in their new roles compared to their old ones. If LeBron James played baseball and Shohei Ohtani played basketball, each athlete would increase their productivity by swapping jobs. As a result each would see an increase in their incomes. Such an outcome would align with a positive relationship between productivity and income. 

However, most companies hire based on unstructured interviews, which lack much evidence for their efficacy. Are we really properly matching workers with questions like “How many barber shops are in the United States? 

Maybe the matching occurs across time rather than across workers. People generally leave when they find their current role unsatisfying, or they believe they’ve accomplished most of what they can from their current role. Yet, when leaving due to dissatisfaction, the worker lowerers their negotiating leverage, so I wouldn't expect a large salary increase from leaving an unsatisfying job. Maybe companies can’t differentiate the satisfied job-hoppers from the unsatisfied ones? They could just assume that every candidate is unsatisfied, and they wouldn’t be far from the truth.

Higher Productivity Firm

Another possibility is that workers tend to move to firms where their labor will be more productive. There’s no mystery why programmers earn more when they move from a no-name firm to Google: Google makes more money. Thus, job hoppers might earn more since they leave to more productive firms.

I still find this unsatisfying. First, why didn’t the worker just start at the more productive firm? Maybe high-productivity companies require experience at the less productive ones. This sounds plausible, and I don’t have any data to confirm or deny it. However, it just doesn’t match with my personal experience. Coding educator Clement Mihailescu began his programming career at Google. Every time I’ve changed jobs, I’ve met people at every stage of their careers. I’ve also seen people on both sides of the job-hopping premium at each firm. It seems like there’s a salary increase from leaving damn-near every company, so I don’t think the presence of high-quality firms explains this. Finally, why wouldn’t these high quality firms just promote from within?

Power Dynamics

Companies probably don’t want to create a culture where people can demand high pay raises and receive them. A firm may refuse a mutually-beneficial pay raise if such a refusal prevents other employees from demanding similar raises. Basically, companies don’t want their workers to feel empowered. But, wouldn’t workers demanding the raise and then quitting leave the firm even less empowered? I just don’t see this as a good strategy for keeping your workers down.

Principal-Agent Problem

A “principal-agent problem” occurs when a representative for an organization undermines that organization in a self interested way. I see a ton of this in data science: employees want to work on projects that interest them or make for good resume bullet points instead the projects that will maximize profit for the company. We might see a similar dynamic in the HR department. If an important employee leaves for another firm, the company itself might lose money, but the people in charge of compensation won’t. In other words, the HR bureaucracy lacks the proper incentives to maintain employees. I don’t think this explanation works for a couple of reasons. First, the same could be said for granting the pay raise. It’s not your money, why not raise your employee’s salary to 11 gazillion dollars? Second, in my experience, executives do judge HR departments for employee churn, so I don’t think there’s any incentive misalignment. 

Compensated Differential

Maybe paying someone to quit costs extra. Employees like working for their current firm or, at least, don’t like the risk of leaving to a new one. An employee there wouldn’t leave for a slight pay raise. As a result, the new employer needs to offer a much higher salary to compensate for the employee’s disutility of switching jobs. I think this explains some of the gap. Few people would switch positions for a 5% raise. However, the raises just seem too high for an explanation to hold. An article linked above says that employees gain 3% for staying and 20% for leaving. I’m not going to fact check that blog, but, in my experience, I hear figures much higher than 20%. Couldn’t the new company provide like, only 12% more? Why is the gap so large?

Also, the compensated differential argument assumes a strong preference for the status quo. That definitely true for risk-averse losers like me, but I don’t think it’s universal. I doubt that most people like their job enough to need a high compensated differential. Everyone thinks their current company sucks. I’ve met people who would probably take less money for the thrill or relief of trying a new role. 

Sampling Bias

In major league baseball, the vast majority of pitchers can’t hit. However, I don’t think throwing a baseball prevents one from hitting it. Rather, I imagine this fact derives from sampling. A pitcher who can’t hit will still make The Show; a left-fielder who can’t hit won’t. In this context, maybe the “leavers” are just more productive workers than “stayers.” I see two problems with this explanation. As mentioned before, companies hire based on dubious unstructured interviews. I don’t see how companies can gauge productivity from these. Secondly, are there really any “stayers?” People job hop all the time. If you’re in a white collar position, browse your LinkedIn connections and try to find a “stayer.”

Psychology

Maybe employers prefer the hot new thing over the tried and true. I don’t find this plausible. First of all, I think we should display skepticism towards explanations that attribute large socioeconomic phenomena to small psychological factors. Many behavioral economics treatments, for example, show pretty negligible effect sizes. Even if we want to consult behavioral economics, it tells us that people prefer the old stuff over the new stuff. Also, think about your company. Does leadership love new things? Or do they prefer doing the same shit over and over? Lastly, if companies prefer new blood, why don’t they fire their current employees more frequently?

Short-Termism

In recent years, a major critique I’ve seen of capitalism concerns short-termism. Companies focus on the next quarterly report instead of the products they can build in subsequent years. Because of this, a firm may not care about the long-term value forfeited from a quit. An immediate raise, on the other hand, would hurt the company right now. But, wait, doesn’t the employee leaving also hurt in the short term? And what about the new firm? Doesn’t offering a large salary to a new employee immediately reduce profits? 

Employees Don’t like Asking for Raises

Another possibility could be that employees don’t feel comfortable asking for raises at their current job, so they can only get one from another company. Workers may feel that asking for a raise may hurt their perception within the company. This just doesn’t match personal experience. People do ask for raises; they just don’t get them. I agree that workers don’t like asking for them, but we hate interviewing even more.

Combination 

Maybe a bunch of “kindas” sum to a “quite a lot.” Each of these factors could influence the job-hopping premium a little bit and, together, they add up to a large raise. This is a bit of a cop-out, but I think a lot of economics works this way. 

Phone a Friend

What do you think? Does the job-hopping premium seem strange to you? Did I dismiss one of these explanations unfairly? Are there a couple (or a couple hundred) that I missed? I honestly don’t know, so I’d appreciate some input.

Also, what does everyone think of open-ended content like this? Don’t tell anyone, but there’s a lot more stuff I don’t know than stuff I do know. I find it fun, sometimes, to consider a bunch of ideas without a discrete conclusion.