Compensation Rules Everything Around Me— How to decide on compensation and what really matters
Note: Republishing this post I shared on the Medium paid subscription a few years ago. This was originally written in February 2015.
Compensation
Compensation appears simple at first brush: what you get for what you do. However, there some interesting and tricky ideas around it. What we get in exchange for our work is often less important than how we get it, when we get it, and why we think we’ve gotten it.
Compensation as Incentive
Most forms of compensation include a layer of incentives, meant to increase productivity. These incentives are often incredibly powerful, and are the unnoticed force behind so many of the challenges of our age. It’s easy to forget that a compensation system is the most simple proxy (and therefore the most often-used proxy) for the value system of the organization.
Charlie Munger, the wisest man I know of, says this of Incentives…
Almost everyone thinks he fully recognizes how important incentives and disincentives are in changing cognition and behavior But this is not often so. For instance, I think I’ve been in the top 5% of my age cohort almost all my adult life in understanding the power of incentives and yet I’ve always underestimated that power.
If Charlie Munger is underestimating the power of incentives, it’s a certainty that we are. One of Munger’s favorite examples is how compensation structure was the innovation that first enabled FedEx to offer overnight delivery. In this case, a small incentive misalignment was the barrier to the success of what is now a ~$50B company.
The integrity of the Federal Express system requires that all packages be shifted rapidly among airplanes in one central airport each night. And the system has no integrity for the customers if the night work shift can’t complete it’s assignment fast. FedEx had one hell of a time getting the night shift to do the right things. They tried moral suation. They tried everything in the world without luck.
Finally, someone got the happy thought that it was foolish to pay the night shift by the hour when what the employer wanted was not maximized billable hours of employee service but fault-free, rapid performance of a particular task. Maybe, this person thought, if they paid the employees per shift and let all the night shift employees go home when all the planes were loaded, the system would work better. And that solution worked.
One small change in compensation, one huge result in output.
‘Surge’ Pricing — Real-time Wage Changes
Anyone familiar with paying Lyft or Uber double the normal rates during rush hour has experienced dynamic pricing dependent on demand. Drivers make more money to drive when passengers place a higher value on rides. It’s worth wondering why other businesses don’t follow this same model. McDonald’s workers make the same hourly rate whether they serve 10 customers or 1,000. Why not reward employees more handsomely when their efforts are more appreciated by customers and more beneficial to the company.
Actually, Zappos has done just that! This article suggested by Tim Cigelske, Director of Social Media at Marquette, explains how Zappos conceived and enacted a system that pays their Customer Support staff more during periods of higher contact volume. The staff loved the freedom it allowed:
Kelly says employees used demand to dictate their schedules but they’re also accounting for personal obligations. “Worker testimonials said it was a lifesaver. They could go home for an emergency and make that time up later in the week,”
Wal-mart: The Single Biggest Regret
Sam Walton of Wal-mart encountered fascinating compensation challenges while growing his behemoth of a mom-and-pop business. While being publicly and enthusiastically criticized for his low wages to in-store workers, he was brilliantly determined to tie profits on a per-store level to manager compensation all the way up his chain of command.
In his book, Made in America, he reflects on some of those earlier choices and shows regret for the way he treated in-store employees. We can all learn from his hindsight observations:
We didn’t pay much. It wasn’t that I was intentionally heartless. I wanted everyone to do well for themselves. It’s just that in my very early days in the business, I was so doggoned competitive, and so determined to do well, that I was blind to the most basic truth, really the principle that later became the foundation of Wal-Mart’s success.
You see, no matter how you slice it in the retail business, payroll is one of the most important parts of overhead and overhead is one of the most crucial things you have to fight to maintain you profit margin. That was true then, and it’s still true today. Back then, though, I was so obsessed with turning in a profit margin of 6 percent or higher that I ignored some of the basic needs of our people, and I feel bad about it.
The larger truth that I failed to see turned out to be another one of those paradoxes: The more you share profits with your associates— whether it’s in salaries of incentives or bonuses or stock discounts — the more profits will accrue to the company. Why? Because the way management treats the associates is exactly how the associates will then treat the customers. And if the associates treat the customers well, the customers will return again and again, and that’s where the real profit in this business lies.
I didn’t catch on to that idea for quite a while. In fact, the biggest single regret in my whole business career is that we didn’t include our associates in the initial, manager-only profit-sharing plan when we took the company public in 1970.
Sam Walton’s biggest regret in his whole, 50+ year business career. Wow.
His point about the paradox is extremely curious. It always seems that a paradox is truth’s way of making us work for it, and this is a great example in the world of compensation.
Incorrect Incentives can Destroy Companies
Incentives are never perfect. There is no perfect system, at best we have an asymptotic approach to aligned incentives. In any set of incentives, there is always a corner to cut, a score to game, or some metric to inflate. If the gap between incentivized behavior and desired behavior is large, it can cripple an organization. Wise words from W. Edwards Deming (who is a fascinating dude and from whom I cannot wait to learn much more).
“If you give a manager a numerical target, he’ll make it even if he has to destroy the company in the process.”
A simple incentive system can easily be ‘gamed’, or optimized for, at the cost of other important dynamics. For example, increasing sales is extremely easy if customer acquisition cost is allowed to rise unchecked. You will end up with lots of sales and no profits.
By pairing opposing metrics and measuring each in context of the other, we can maintain a healthy increase in the target metric. Decrease the fraud rate and maintain or decrease the false-positive rate. Increase sales and increase or maintain profit margin. Decrease customer acquisition cost without lowering customer life-time value.
Using these competing metrics can help keep incentives focused and prevent overzealous managers from getting tunnel-vision and unknowingly (or knowingly, if greedy/stupid) destroying your companies’ value.
This idea is explained thoroughly by Keith Rabois in his lecture from How to Start a Startup, and was suggested here by Max Olson of Futureblind.
Max Olson also shared some of his notes from The Success Equation. These explain how to assess which metrics should be evaluated for inclusion into incentive programs.
Useful statistics have two features. (1) Persistence, which means what happens in the present is similar to what happens in the past. This is also called reliability — if luck is more important, than you would expect the reliability to be low. (2)Predictiveness of the goal you seek. This is also called validity, meaning it is valid to conclude that one measurement causes another. It compares two values. In basketball, an example would be percentage of shots made and total points scored.
Statisticians assess persistence and predictive value by examining the coefficient of correlation. The process of determining which statistics are useful begins with a definition of your objective. Knowing your objective is important because it’s hard to chart a course without knowing the destination. Next, you have to determine what factors contribute to achieving your objective.
You can plot a measurement on a chart that has the luck-skill continuum as the horizontal axis and the predictive value on the vertical axis. The most predictive and persistent measurements are best. (Examples include on-base percentage for baseball, or active share for investing.)
Subtle Dissenters on the Power of Incentives
Despite anecdotal evidence about how important incentives are to efficient working environments, and pervasive use of incentives in our compensation plans, there is some academic evidence to the contrary.
Alfie Kohn writes (for the Harvard Business Review) a piece that is unambiguously titled “Why Incentive Plans Cannot Work”. Interestingly, he makes the point that our understanding about the research that our plans are based around is vastly oversimplified, and dangerously shallow.
The findings suggest that the failure of any given incentive program is due less to a glitch in that program than to the inadequacy of the psychological assumptions that ground all such plans.
[…]
Do rewards work? The answer depends on what we mean by “work.” Research suggests that, by and large, rewards succeed at securing one thing only: temporary compliance. When it comes to producing lasting change in attitudes and behavior, however, rewards, like punishment, are strikingly ineffective.
In our workplaces, we need more than temporary compliance. We need systems that keep us engaged and interested for our entire lifetimes, if we hope to keep our work above the status of necessary drudgery.
Kohn has a fascinating point about reward and punishment. We often hear that rewards (positive reinforcement, a la BF Skinner) are effective where as punishment (negative reinforcement) is not. Kohn says that these eventually feel identical to the subject:
Punishment and rewards are two sides of the same coin. Rewards have a punitive effect because they, like outright punishment, are manipulative. “Do this and you’ll get that” is not really very different from “Do this or here’s what will happen to you.” In the case of incentives, the reward itself may be highly desired; but by making that bonus contingent on certain behaviors, managers manipulate their subordinates, and that experience of being controlled is likely to assume a punitive quality over time.
This is a point that is hard to stomach, as few managers want to feel as though they’re punishing someone. Everyone wants to be the good guy. Kohn teaches us that bestowing (or withholding) rewards are not the way to adoration.
For whatever reason, (social pressure, habit, consistency bias, or just bureaucracy) compensation practices don’t seem to have changed very much in the past few decades. In his article, Kohn mentions that for nearly 40 years, academics have been debunking the compensation best practices of the private sector.
The funniest part about Kohn’s piece? It was written 20 years ago. So let’s make it 60 years.
Drive: The Surprising Truth of What Motivates Us
Dan Pink is a recent popularizer of many of academia’s debunking mentioned by Kohn. You may have seen his TED Talk or read his book, Drive, which was suggested as a great resource on Compensation by both Bo Fishback and Max Olson.
In Drive, Pink allows for two kinds of tasks, Algorithmic (following set instructions) or Heuristic (creative or requiring a novel solution). Incentives (rewards & punishment) can work nicely for Algorithmic tasks. However, here is the surprising part—they’re massively detrimental to people working on Heuristic tasks. Essentially, if creative workers must be intrinsically motivated, or they’re extremely unlikely to produce good work.
His TED Talk lays out the basic three requirements for someone to feel intrinsically motivated: Mastery, Autonomy, and Purpose. They have to feel that they’re learning and mastering a task, that they have the freedom to make their own mistakes, and that their efforts are going to a greater good.
Furthermore, it’s worth it to make the extra effort to attract and retain the intrinsically motivated; they almost always achieve more than shorter-sighted reward seekers.
Highly recommend Dan Pink’s book, Drive based on some fantastic references from contributors. If you’re more into summaries, this is a good outline of the book.
Some Things We Didn’t Think About
A book called True Professionalism by David H. Maister has some valuable gems buried in it about compensation and the effects is has as an incentive. While this book is specifically focused on Professional Service Firms (legal, accounting, etc.) it has many bright ideas from that discipline that are applicable and important for any business.
On the use of conditional rewards (like bonuses) for individuals:
In a firm which makes extensive use of individual performance rewards, it its often the case that the firm’s reaction to a mediocre or average performer is to say “That’s ok, you can stay — we’ll just pay you less.” This is hardly a recipe for excellence!
If you’re trying to run a world-class organization, merely saving some money by withholding some pay from under-performing employees is going to be extremely counterproductive. Investing in their growth (or their replacement) is a far more productive set of options.
Firms with individual performance-based reward systems often end up tolerating wide varieties of performance, while those with more group-reward systems are usually less tolerant of performance problems.
We probably don’t want to be in the business of ‘tolerating wide varieties of performance’. Maister mentions how results are improved when group or firm-wide performance rewards are instituted.
It is a fascinating paradox that many high-performing firms in a wide variety of professions make minimal use — if any — of individual performance results in sharing firm profits. Instead, they use pay schemes that give significant weight to firm-wide or group-wide results.
This is counterintuitive, as the economic assumption would be that the more dissociation there is from an individual, the less effort each individual would contribute. However, this proves to encourage teamwork, support, and growth in order to improve the entire team or companies’ performance.
One more subtle and very important distinction from True Professionalism:
To create a great firm, management must run a system which causes performance to improve, not one that simply rewards improved performance where it happens to manifest itself.
True Professionalism is a fantastic book and a must-read for anyone in Professional Services. Those of us who don’t have to wear suits to work could learn a thing or two from it too.
Perception of Compensation vs. Reality
Compensation, like everything else, is subject to our perceptions. Some people care more about stability, some about upside, some about title or opportunity or perks. How we understand and value our compensation affects our (and our employee’s) happiness and efforts at work.
Relative vs. Absolute Pay
Numerous studies have taken a look at the psychology behind compensation. Interestingly, people most commonly use relative measures (compensation of peers) to determine how they value their pay. Here’s one (dense) academic paper published in the Journal of Economic Behavior & Organization.
Frank (1985a) declares that “someone whose close associates all earn $50,000 a year is likely to feel actively dissatisfied with his material standard of living if his own salary is only $40,000… Yet that same person would likely be content if his closest associates earned not $50,000 but $30,000 a year.’’
If they’re making more than the people around them, they’re happy. If they’re making less, they’re unhappy. It’s not the absolute amount that matters, but the relative amount. This is pure perception.
Andy Grove gives some nuance to this concept in High Output Management—since obviously for some levels of income (or more accurately, income/expense ratios), the absolute amount is certainly the more important piece. It’s extremely helpful to know which aspect of the compensation an employee values, relative (perception) or absolute (purchasing power).
If the absolute amount of a raise in salary is important, that person is probably motivated by the physiological or safety/security needs. If the relative amount of a raise—what they got compared to others — is the important issue, that person is likely to be motivated by self-actualization, because money here is a measure, not a necessity.
Income Volatility is Not Income Risk
Some forms of compensation are highly variable—total income can change hourly (traders) or daily (plumbers). However, this variability does not necessarily make it risky.
Much more risky than earning a slightly variable amount each day is the risk of, at any moment, suddenly making $0. This is life on a salary.
Here is an excerpt from Antifragile explaining the antifragility of the artisan and the extreme Fragility of the salary worker. It’s crucial to not assume that a salary is safe because it lacks variability—it could disappear any moment, and disappear completely.
Artisans, say, taxi drivers, prostitutes, (a very, very old profession), carpenters, plumbers, tailors, and dentists, have some volatility in their income but they are rather robust to a minor professional Black Swan, one that would bring their income to a complete halt. Their risks are visible. Not so with employees who have no volatility, but can be surprised to see their income going to zero after a phone call from the personnel department. Employee’s risks are hidden.
[…]
[The Taxi Driver] has the freedom to continue until he drops (many people continue to drive cabs into their 80s, mostly to kill time), since he is his own boss, compared to his [Peer on Corporate Salary], who is completely unhireable in his fifties.
Let’s not mistake lack of volatility for lack of risk, and believe in the safety of a paycheck that could drop off at any moment.
Antifragile is a difficult book to excerpt from because of it’s carefully constructed ideas, so if this makes no sense, I encourage you to read the book. It’s truly fantastic.
You Cost more than Your Compensation
Employees tend to believe that if they’re making $70,000 per year, that’s what they cost the company, and a $70,001 contribution would make a profit for the company.
In fact, companies take on many expenses that increase the cost of hiring to 1.25–1.5x the cost of the salary. Before the employee ever starts, the costs of recruiting and onboarding have been incurred. As soon as they do start, taxes and insurance costs begin: Social Security Tax, Medicare Tax, and State Unemployment Insurance, not to mention Health Insurance Benefits, 401(k) benefits, and other perks.
For someone earning a $70,000 salary, the company is likely incurring a cost close to $90,000 per year for that employee’s time—not including overhead required to make that time productive like space, equipment, etc.
It’s important as an employer to understand that you must be able to make a profit after an employee’s total cost, not just their hourly or yearly pay. Also, it may be helpful to be transparent with employees about the total costs that the business incurs on their behalf.
Here is some interesting additional reading about this in a Column from MIT, and a short piece on CNN Money.
Compensation as Purchasing Power
No one is working in order to earn cash and build an igloo out of it. We work to earn money to exchange for goods and services to make our lives better. What companies are really giving people as compensation is purchasing power.
By hiring employees in cheaper areas of the country (or the world), companies can trade the same purchasing power for more of the employee’s effort. This is something that Jason Fried mentions in his book, Remote. Through Odesk and Elance, it’s easier than ever to discover and leverage talent outside of your geographic area.
Thanks to Andrew Farah for sharing this idea about compensation and sharing Remote as a resource to learn more.
Brave New World of Salary Transparency
In the frontier of companies holding transparency as a company value in Startup Land, Buffer seems to be in the lead when it comes to Transparency about Compensation. Not only is it public within the company, but it’s also public to the… public. Here’s what everyone at Buffer makes (and an awesome post about how and why).
This plan has some brilliance to it. The relativity problem is out the window within the team—everyone knows before they even start where they stack up and why, and if they don’t like it they don’t join. There’s no possible scandal when it’s this open.
Keith Rabois mentions that Steve Jobs tried this during his tenure at NeXT as well. Rabois also counters, in response to a common critique of this idea, that professional sports have extremely transparent compensation, and they don’t seem to suffer any related teamwork issues. Moreover, since we know that they’re looking for self-actualization from their income (remember Andy Grove’s types of employees?), the relative compensation should be more important to them.
Thanks to Daniel Pearson and Blake Robbins for both suggesting this Buffer Post about their open salaries.
One Tiny Note from Ben Horowitz
In his incredible book The Hard Thing about Hard Things, he makes a small note of immense importance on Compensation.
Compensation has the potential of becoming an extremely political issue. To avoid allowing a slide into chaos, Horowitz recommends building strict processes and never deviating. This means set timelines for possible raises, based on clearly pre-determined criteria, and do not entertain requests outside of your processes.
He explains why in a story with musical accompaniment at 8:37 in this lecture at Stanford for How to Start a Startup. Important to watch.
Types of Compensation
This may seem unnecessary, but if even one item on this list surprises you, it was worth reading. There are many types of compensation, certainly more than will show up here. Having a broad and amorphous understanding of the concept of ‘compensation’ allows us creative opportunities. Learning how employees value different types of compensation allow for chances to make arrangements that benefit both employees and employers.
Salary
Curiously poor in terms of incentives. No variability and all-or-nothing.
Check out this tool from Wealthfront to see Startup Salary/Equity compensation make-up for different verticals and experience levels. Thanks to Jason Evanish for the link!
Also, Amazon has an interesting policy of Salary Caps. Salary is just to give people living money (into the self-actualization side), anyone getting rich at Amazon is doing it from stock, not from salary. Thanks to Bo Fishback for the tip and the link here!
Hourly
The simplest method of trading time for compensation. Can quickly devolve into minimum effort in order to not get fired.
Shift
Employed to brilliant effect by FedEx in our earlier story. Effective for speed-related work (probably an acceptable accuracy/quality threshold).
Piecework
Compensation contingent on volume of completed work of acceptable quality. Encourages innovation and hard work—good incentive alignment if quality is maintained.
Commission
Sales is a beautiful place in the compensation world—simple enough to create direct incentives for. Commission can be dangerous without tight controls around treatment of clients, so balance there is important. Also, some commission structures reward actions that would have happened whether the salesperson was there or not—an interesting challenge for a business.
Performance Bonus
Possible in different forms, general category of reward contingent on a goal, which we covered pretty thoroughly in the Incentives section.
Stock Options, Grants, Discounts, and More
Designed to give employees upside in the success of the company without having to put any capital at risk, stock options are common in startups, and some public companies (in different form).
This isn’t supposed to be a deep dive into Stock compensation, (maybe another time) it’s a deep topic with some interesting takes on it recently.
The Pros and Cons of owning stock in your employer is a self-contained debate on if (and how much) to invest in the company you work for.
Here’s Charlie Munger on Stock Options (Thanks to Max Olson!):
As a mechanism for compensation, a stock option plan is capricious, as individuals awarded options in a particular year would ultimately receive too much or too little compensation for reasons unrelated to their performance. Such variations could cause undesirable effects, as participants receive different results for options awarded in different years. In addition, a conventional stock option plan would fail to properly weigh the disadvantage to shareholders through dilution…
Also, Sam Altman of Ycombinator wrote an interesting post about the weaknesses of Stock Options as compensation for Startup Employees. Ben Horowitz’ rebuttal and discussion of this is in his presentation at Stanford.
Thanks to Max Olson and Andrew Farah for both suggesting Altman’s excellent post on Employee Equity.
I’ve written more about stock options in my article entitled, Stock Options are worth less than you think.
Perks
401(k) matching, health benefits, free meals, gym memberships and company Lyft credits all fit in here, and some perks can be considerable. For ridiculously profitable companies like Google, it’s sensible to keep costs inside the company and create lavish perks to retain talent and improve productivity.
Small business owners often have the luxury of the same on a smaller scale, buying company cars or meals.
Intangible Types of Compensation
Not everything you’re getting for your work can be measured and compared tangibly. The more important aspects of your compensation aren’t on your offer letter, they’re in your company culture and team.
Learning
Does this job offer chances to learn new things that are important and helpful? As Dan Pink would say—Mastery?
Mentorship
The most important factor about a job is likely the manager and the team. These must be people that welcome, support, teach, and empower.
I’ve heard a story about an employee joining a company with a CEO that she admired who gave back $10,000 in salary in exchange for weekly dinners with the CEO to ask questions and actively learn from them.
Purpose
Is the company based on a business model no better than blackmail? Pushing unnecessary drugs? Marketing addicting foods? Practicing predatory lending? Hard-selling dubious financial services?
If you don’t feel good about what your company does, you’re taking on some burden of existential suffering. Get the F**k out. […] Yes, you can.
Maybe your organization’s mission is so important and benevolent that it is fulfilling by itself, and the paycheck is just the bonus that helps you eat while you do good things with your life. Rock on.
Opportunity
For any dynamic employee, it’s important that there is a path forward. It doesn’t necessarily have to be obvious, paved, or cleared, but it must exist.
It’s important to realize that if you’re not getting one of these types of compensation, you‘re being compensated to forego it. It may be helpful to consider these types of compensation when offering a job or accepting one.
How Compensation is Determined
The most difficult part, of course, is putting all of this together into an ingenious arrangement of aligned incentives with intrinsic motivation, perfectly communicated, deliberated through a strict and well-defined process in such a way that the company benefits and increases in value.
Unfortunately, these stories are hard to find. Andy Grove puts it well:
Social norms can force us into some unfortunate compensation practices.
In his book, High Output Management, Grove explains the two main levers for compensation, Experience and Merit.
Unions and most government jobs lean toward pure experience-only salary scales. Apart from whether this is fair or not, the message from management is that performance doesn’t matter much. Consider teachers in many school systems. A good one gets paid the same salary as a bad one if they both have been around for the same length of time.
This extremeness is balanced on the other side by merit-only compensation, such as 100% commission jobs. Most jobs settle somewhere on an unhappy compromise between the two extremes of merit-only and experience-only.
Another characteristic of employees which figures into their compensation is Education. Degree inflation makes a college degree required for many jobs now, and compensates employees for advanced degrees.
Thanks to Tom Janiak for submitting the article on college degree inflation.
Proxies for Compensation Decisions
Most compensation offers are based on ‘Market Value’ which refers to a giant spreadsheet full of data on what other comparable people in the workforce are making.
This fantastic story about a job offer, negotiation, and final offer is from Michelle Wetzler (who now works at Keen). She wrote about the entire process and how she did her own research, used multiple models to estimate her value to the company, and counter-offered at a higher rate. There’s a lot of lessons in Michelle’s simple story, from the rigor of her analysis to the chutzpah of her request.
Paul Graham wrote The Equity Equation, to answer these questions:
An investor wants to give you money for a certain percentage of your startup. Should you take it? You’re about to hire your first employee. How much stock should you give him?
He provides a formula that tests for the increase of the average outcome. If a hire improves the average outcome of the company, net the equity and salary they’re being given, then it’s a good decision and should be hired.
Thanks to Alex Blunk of Fort Point Brewery for suggesting both of these posts.
Final Words of Wisdom
To close out, a few quotes that either have to do with compensation, or could do well to be applied to the challenges it presents.
“Perspective-shifting is much more powerful than being smarter.”
“Generosity is it’s own form of power.”
“You’re either lucky to work here, or we’re lucky to have you.”
“The three most harmful addictions: heroin, carbohydrates, & a monthly salary”