Research Brief

Google Says to Hire the Best, but Suppose You Can’t?

Top companies hog the best talent, adding to their competitive advantage

It’s not often that a human resources guy gains corporate guru status, but former Googler (his term) Lazslo Bock is a notable exception. His 2015 book, Work Rules! Insights from Inside Google that Will Transform How You Live and Lead, has taken on biblical status among executives who hope to hire and manage just like the wildly successful search giant.

Bock’s core advice: Hire great talent, not people you have to train to be great. “The presence of a huge training budget is not evidence that you’re investing in your people,” he writes. “It’s evidence that you failed to hire the right people to begin with.”

Sounds like brilliant advice, until one stops to think that, by the time Google and Facebook and Apple and a handful of other dominant and deep-pocketed companies have sorted through the available talent, there might not be so much greatness left to hire. That’s what occurred to UCLA Professor of Economics Simon Board, who decided to research how the hiring practices of the Googles of the world affect less dominant firms.

Opt In to the Review Monthly Email Update.

“Not every firm can hire the best,” Board said in a phone interview. “Some firms have to hire the average.”

working paper by Board and UCLA Economics colleagues Moritz Meyer-ter-Vehn and Thomasz Sadzik suggests the recruitment prowess of dominant companies causes most of their industry to fall further and further behind. “It’s the case of something that is beneficial to a single firm being harmful to an entire industry,” Board says.

While “hiring the best” greatly benefits Google — as well as employers like elite universities, top hospitals and others that dominate their fields — most competitors can’t replicate the strategy, the study indicates. As the years go on, the leaders’ continual cherry-picking of top talent increasingly disadvantages everyone else, including job seekers who narrowly miss out on being hired by a company like Google, the study’s model suggests; these almost-hireds end up back in the candidate pool for lesser firms, undistinguished from lesser candidates.

Applying a game-theoretic model, the study finds that wage and productivity inequities within an industry compound when leading companies attract the best talent. The great talents attracted by dominant firms in turn become great recruiters of still more talent for those firms, and the rest of the industry perpetually lags.

“Talent,” the study finds, “is a source of sustainable competitive advantage.”

An Improbable Fix

The model was built to reflect the diversity of talent typical of any industry, be it musical theater or technology. The recruiters include organizations that pay top salaries and are adept at spotting good hires, as well as those that hire randomly. The pool of job applicants includes extremely talented individuals alongside those incapable of the jobs.

Ideally, for both employers and job seekers, unqualified applicants would drop out of the pool entirely to pursue different lines of work, Board explains. But in the current system of recruiting, they simply move on to the next company to compete once again against stronger applicants.

In the model, the high-paying, savvy-selecting Googles of industries land every applicant they want. Like all employers, they make some hiring mistakes; great candidates get overlooked, bad fits get hired. But their odds of hiring a top talent are inherently better than others because the job pool is still, let’s say, half full of good candidates.

Once Google, Facebook, Apple and the like take their pick of the market, the share of good candidates drops to perhaps 40%. Facing an applicant pool that now has a higher percent of unattractive candidates, the next firm down is likelier to hire badly. If that firm takes a better qualified candidate instead, the next recruiter is even likelier to make a hire it will regret.

This negative fallout does not so neatly occur when non-leading companies — say, firms like one we’ll call Acme — offer high wages and take their pick before the dominant firms do, according to the model. The less skilled recruiters at hypothetical Acme make more good hires when they’re able to pay up and get first crack at the talent pool, but they still make enough mistakes to leave behind plenty of talent for the industry leaders, Board explains.

“It’s like a river with lots of fish,” Board says. “You really don’t want to be downstream from a really good fisherman (Google) who picks out the best fish. But a bad fisherman (Acme) upstream won’t matter.”

While this Acme-goes-first scenario is unlikely in real life, companies might consider investing more in recruiting and salaries to attract more talent. But these non-dominant players have very little chance of ever stealing away the best recruits from Google-level players, Board says. The top companies can always top whatever wage or perk their competitors offer.

“Whatever Acme is willing to do, Google is willing to do more,” he says. Acme might occasionally make an offer good enough to get a Google pick, he says, but that’s not a sustainable long-term strategy.

Meanwhile, the stronger hires are helping the dominant companies leap ahead of the lesser competition, according to a large body of research. For example, a salesperson in the top quartile of producers is on average more than 1,000% more productive than the 25th percentile salesperson, according to a study published in the Quarterly Journal of Economics.

If Information Were Free

If all the industry players below the top tier suffer from this situation, it suggests the industry overall underperforms. But suppose recruiting aimed at industry-wide efficiency. Google’s passed-over, unqualified applicant, “Bob,” would come with a dossier on everything Google privately learned about him. Google, like most dominant companies, is exceedingly good at using employee referrals, interviews, job tests and other internally produced information to extract the best of the labor pool and send others back into the job market, according to the study.

With Google’s information, second- or third-tier Acme would learn a little more in its interactions with Bob, which it would in turn send on to the next company that considers him. The collective wisdom of recruiters would quickly place great talents where they fit and send the absolute laggards off to other careers, Board explains.

But of course, Google would prefer its competitor hire an incompetent Bob, and so would Acme.

“The way the competition works, the firms aren’t helping each other at all,” Board says. “They’re just hurting each other.” Bob will either land a job and drag down his employer, or he’ll be unemployed until he exhausts the industry possibilities.

In our current system, not much disqualifies Bob. A felony record would likely weed out Bob from this job pool very quickly, which, regardless of fairness, is optimal from an economic perspective if he’s going to be weeded out eventually. Industry wages and employment are higher when there is such “conclusive” information available, such as results of drug tests or college performance, the study finds. These details make employers more confident about candidates and likelier to hire, Board says. (“Inconclusive” information like credit scores, however, may lower wages and employment, according to the study.)

These findings address an unsettled area of research into “ban the box” laws that outlaw job application check boxes requiring candidates to reveal criminal convictions in the initial screening. Some research, including a 2019 study in the Journal of Labor Economics, suggests the laws decrease minority hirings because employers, lacking actual data on criminal backgrounds, speciously use race as a proxy for criminal histories.

Overall, the conclusions from the UCLA paper suggest that Google’s recruiting advice is spot on: hire the best for long-term success. Realistically, it finds, few firms will be able to follow it.

Featured Faculty

About the Research

Board, S., Meyer-ter-Vehn, M., & Sadzik, T. (2019). Recruiting talent.

Related Articles

A cube of hundred dollar bills wrapped in a blue bow Research Brief / Productivity

Trying Out Bonus-Pay Theory on Unsupervised, Low-Skill Tasks

Incentives boost output, but benefits level off at a fairly low point

Forklift in a warehouse Research Brief / Manufacturing

When Manufacturers Commence Exporting, They Become More Efficient

Looking at costs, in a sample of 5,000 plants in Chile, remarkable productivity gains occur