The Motivation Toolkit by David Kreps
The Motivation Toolkit by David Kreps
The Motivation Toolkit by David Kreps
KREPS
The
Motivation
Toolkit
■■■
Motivation = Alignment
The traditional image of the relationship between a supervisor and
subordinate (or employer and employee) is that the boss tells her
subordinates what to do and what not to do, and monitors their compliance
with these dos and don’ts, with dire consequences in case of violations.
With Type-K jobs, this method of “control” is not going to work well. If
you, the boss, can’t say up front what your employees should do beyond “be
creative” or “cooperate,” and if you can’t tell after the fact how well or how
hard they tried to fulfill your commands, dos and don’ts alone probably
won’t buy you more than perfunctory performance.
And, in fact, dos and don’ts often provoke behavior that you don’t want.
People tend to resent limitations or rules imposed on them and actively
attempt to subvert such limitations and rules. A classic dramatic storyline
involves a new resident in a dark and creepy Victorian mansion who is told,
“Whatever you do, you must not go into that room.” This, we all know, means
that the individual will go into that room, with unpleasant consequences. This
might be blamed on curiosity; in the work context, it is probably more often
the psychological “discomfort” that comes from feeling controlled, which is
best mitigated by the employee showing his or her ability to control their
own actions by going against what they are told to do and not do.
Psychologists call this phenomenon reactance.
You, as employer, might think that with cleverly designed rules and
restrictions, you can frustrate the employee’s innate desire to circumvent
them. And there is a sense in which this is so: Psychological research
suggests that reactance is strongest when the individual is most conscious of
the rules and restrictions that are imposed. If you can design the desired rules
and restrictions in ways that don’t engage the employee’s active cognitive
processes, you will in this regard be better off. But this is a tall order. Posing
the rules as “guidelines” may dull reactance, because it gives the employee
the sense that he has some measure of control. But if rules are merely dressed
up as “guidelines,” employees will see through the window dressing. Finally,
designing rules and restrictions that take a lot of employee creativity and
guile to circumvent may seem a good idea. But clever control mechanisms
are apt to engender equally clever attempts to subvert the control. This is not
how you want your employees using their creative energies.
So what is the alternative to direct control; that is, to dos and don’ts?
Robert M. Bass, one of the most successful and distinguished graduates of
Stanford GSB, summarized his HRM strategy as follows:
_____________
*
At least, it did so on October 26, 2016.
2
2. But advocates for cutting the guarantee felt that, if Safelite cut the
guarantee, more technicians would choose to put in enough effort to
be on the piece-rate part of the compensation scheme. With a 100%
guarantee, a technician’s choice is to do 10 windshields per week
and earn $480, or do 17 or more (at the $30 piece rate) to earn
more. But with a 70% guarantee, which is $336 per week, the
same technician only needs to increase his work rate by 20% (to
12 per week) to beat the guarantee. And even if a technician is so
work-averse as to settle for 10 windshields and $336, this gives a
unit labor cost of $33.60 for Safelite, instead of $48 per
windshield with the 100% guarantee. Bottom line: With the lower
guarantee, more technicians will opt for doing more than the
minimum, at a unit cost of $30, and those who don’t choose to do
more than 10 still provide those 10 at lower cost per unit.
■ Finding, upon arrival at the job site, that the glass loaded on his
truck by the warehouse folks is the wrong size or defective, is not
pleasant for a technician. So Safelite, as it implemented PPP,
invested in improving its warehouse operations.
And as the employer thinks through the various factors that employees can’t
control but that affect the outcomes they achieve, the employer should think
what form of “insurance” is most effective for dealing with the risk those
factors impose. Safelite’s guaranteed minimum wage offers insurance, but it
is a fairly blunt insurance instrument.
Suppose that, having improved dispatch and warehouse operations, the
remaining uncontrollable risk for technicians is work starvation, where the
technician is ready and willing to take another job, but no jobs are available.
With a weekly wage guarantee, a technician who has been “starved” on
Monday and Tuesday may decide that it is no longer worth working hard on
the last three days of the week; the odds of earning above the weekly
guarantee are too small. If this is the main remaining risk factor, Safelite
might pay an hourly wage to technicians while they are “involuntarily idled”
rather than offering a guaranteed weekly wage. This scheme, which is
probably the most common way to fix the problem of work starvation in
piece-rate settings, is not without its own problems in the context of Safelite,
especially if the work available is assigned in strategic fashion: A technician
who wants to coast for a few days influences the dispatcher to assign work to
other technicians, as long as other technicians are available. But, all things
considered, it may give a more efficient overall arrangement than a weekly
guarantee.
Suppose you operate a Honda new-car dealership, and you wish to motivate
members of your sales force. You could look at number of cars sold and the
margin achieved in those sales as measures of how well each salesperson
has done. But these measures are noisy reflections of how well a particular
salesman has been selling; for instance, he may have been unlucky in the
draw of customers to whom he was assigned. Particular cars may be difficult
to sell because of some new model brought out by the competition. And when
the local or national economy is doing poorly, consumers may be generally
less inclined to buy new cars.
There isn’t much one can do to about the luck of the draw when it comes
to customers. But the local or national economy for this Honda salesman is
the same as for the nearby Toyota dealership. So by benchmarking the
performance of a Honda salesman against the overall sales of Toyotas, you
can eliminate this sort of “noise,” at least to some extent. By this I mean, if
sales of Toyotas fall off, but a Honda salesman maintains his level of sales
and margins, this indicates excellent (relative) performance by the Honda
salesman, so his incentive reward should be higher. Or, put a bit more
negatively, if you see sales of Toyotas fall off, you should be less inclined to
blame a similar fall-off in the performance of your Honda salesman on his
lack of sales effort, and his incentive compensation should be adjusted
accordingly.
But what if Toyota introduces a new model that sells like hotcakes and,
accordingly, depresses the sales of a competing Honda model? A Honda
salesman is not responsible for this decline in sales, and benchmarking
against Toyota sales compounds the “noise.” The best benchmark you can
find for a specific Honda salesperson is then the performance of other Honda
salespersons. They are selling the same vehicles, in the same economy. So
maybe you want to benchmark how one of your salespersons does against the
performance of Honda salespersons nationwide, or statewide, or even in
your own dealership.
Benchmarking the performance of one salesperson in your dealership
against the performance of your other salespersons would seem to eliminate
as much noise as possible. But doing this raises potential problems: If you
want your salespersons to act as a team, with one salesperson helping
another to make a sale, benchmarking each salesperson against the
performance of others introduces adverse incentives; why should
Salesperson A help B, when sales made by B has a negative impact on A’s
compensation? This consideration isn’t dispositive; the natural inclination
that folks have to help one another (what later will be called the norm of
reciprocity) can overcome this adverse incentive effect. But, especially in
cases where cooperation among employees is crucial, you should watch out
for it.
When your benchmark for individual performance is the average
performance of other employees doing the same task, you should also watch
out for peer pressure in reverse; that is, where rate-busters are “encouraged”
by their peers to dial down their efforts. In a similar vein, re-setting piece
rates downward based on enhanced production rates, particularly when the
enhanced production rates are due to implementing a piece-rate scheme,
rarely works out well for the employer.
One extreme form of benchmarking (which, to some extent, is present
whenever you benchmark employees against each other) is called
tournament-style compensation, where you give a prize to the best
performing employee in a group, or to each of the top 5% in terms of
individual performance; think of a case where the salesperson with the best
record in each six-month period gets a bonus of $5,000 or a vacation trip to
Hawaii and the runner-up gets a weekend for two at a local spa.
Tournaments, especially in organizations with a culture of competition, can
be very effective. But they certainly raise issues when it comes to
cooperation among the “contestants.”
Dynamics
Multitasking
The bottom line is that the more complex the job in terms of the number of
tasks and the number of dimensions of quality for those tasks, the less
effective economic incentives can become, because the harder it is to
balance incentives for the different tasks and their dimensions.
Screening Effects
A benefit to Safelite of its incentive scheme was that it not only motivated
individual technicians to work faster, but also differentially motivated
different types of employee and prospective employee, thereby improving the
average quality of the population of technicians employed by Safelite. As
noted before, careful measurement of the impact of the incentive scheme led
to the conclusion that it caused unit labor costs to fall by around 30%, and
fully half of that improvement was due to an improvement in the quality of
technicians employed.
The point is obvious but important: The way you compensate your
employees triggers different reactions in different types of employee, and so
it changes the characteristics of those who choose to work for you and who
stay on the job. Economists call this a screening effect.
Screening effects can be so strong that they can become the basis for a
firm’s business strategy. A great example of this is RE/MAX, the large
American real-estate brokerage firm. Traditionally, realtors working for a
large agency receive a base wage and a fraction of the commissions they
generate from purchase and sale of properties, in return for which the agency
provides offices, clerical assistance, and so forth. RE/MAX turned this on its
head: Agents at RE/MAX can keep 100% of any commissions they generate,
and they pay RE/MAX for the office space, clerical support, and so forth that
they receive.‡ And, in fact, they pay RE/MAX more for these things than they
would have to pay if they bought equivalent services on their own; RE/MAX
profits from the difference in what its agents pay for these services and what
it costs to provide the services.
Why are RE/MAX agents willing to pay this extra? Because, by doing so,
they are labeled as RE/MAX agents, which is a valuable label. Within the
population of all real estate agents, RE/MAX agents are known for having
one or both of two characteristics: they are (a) more talented in their abilities
to make deals, or (b) willing to work harder to make more deals. If you, as a
buyer or seller of a house, want a realtor who is more likely to have qualities
a and b, and you don’t know of a particular agent with these characteristics,
RE/MAX is the place to find one.
The reason the RE/MAX label persists as a strong signal for qualities a
and b is due to the screening effects of the RE/MAX compensation scheme.
This scheme is attractive to agents who know that they have characteristics a
or b or both. It is unattractive to agents who aren’t willing or able to work as
hard or as well. Of course, agents who don’t have characteristics a or b
would still like the label RE/MAX agent, and the signal it provides, but the
price they have to pay—taking the RE/MAX compensation scheme—is too
high a price for them. In other words, the RE/MAX compensation scheme,
through the screening effects it naturally triggers, is the foundation of
RE/MAX’s business strategy.
Recall the conversation about the basic trade-off in incentive theory:
incentive versus risk shielding. RE/MAX goes to the extreme of no risk
shielding at all for its agents. Therefore, besides characteristics a and b,
RE/MAX agents as a group tend to be less risk averse, whether because they
have fewer financial obligations to meet (such as a family to feed) or they
have more financial resources to call upon during a dry spell or simply
because they are less bothered by risks. Having an agent who is more willing
to gamble for the big payoff is not necessarily desirable for the client. So this
is a less desirable aspect of being labeled a RE/MAX agent. But, on balance,
a and b are desirable enough characteristics to be perceived as having, to
overcome this.
Recap: The Economic Theory of Incentives
3. Especially in the very active job market of Silicon Valley, but also
in general, firms looked for schemes that would fix employees
(reduce voluntary turnover), and features of options such as the
loss of as-yet unvested options and the six-months-to-exercise-if-
you-leave features increased the cost to an employee of departing.
B. Top managers of the firm, on the other hand, may through their
efforts on the job have considerable influence on the price of
equity of the firm. For such managers, the risk-incentive trade-off
makes some sense. But (at least) two second-order considerations
arise for these folks.
Suppose the company in the example given a few paragraphs
prior is struggling, with a price of equity around $10. Suppose the
CEO holds a lot of options with a strike price of $15. At an equity
price of $10, these options aren’t worthless: If she (the CEO)
could sell them, she might get something for them, since it is
always possible that the company will recover and its common
stock price will get above $15. But they probably aren’t worth
much. And the act of selling out-of-the-money options is probably
outside the rules (the rules often forbid selling options at all;
either exercise them and sell the shares, or hold them). Even if
permitted, the optics of a CEO selling off the options she holds in
the company she runs are . . . not good. So, with all that non-
wealth tied up in options, what are her incentives? They are, at
least in part, to do what she can to get the price of the stock up
above $15. From her perspective vis-à-vis these options, it
doesn’t matter to her if the price of the company’s shares is $10 or
$5 or $2. Her options are worth nothing in all three cases. So the
options incline her to take gambles that might pay off and get the
price of equity up to, say, $20, even if those gambles probably
mean a price of equity down in the $2 range. This is just the flat-
spots problem already discussed.
On the other hand, suppose top managers of the firm have a lot
of their personal wealth tied to the fortunes of the firm. If they are
personally risk averse, they may forgo taking risky investments
that would be in the interests of their more diversified
shareholders. Options allow them to benefit more from the upside
than they lose on the downside, which might motivate them to take
such risky investments.
And, to complicate matters still further, top management has
many other incentive concerns weighing on them. To assert that
options align their interests with those of the shareholders in the
firm, at least in this sort of situation, is a good deal less than
completely obvious.
_____________
*
At least, that’s what comes out of an economic-theory analysis of this situation. You’ll need to consult
a good economics textbook—one that provides formal models of incentive theory—to see why.
†
I’m assuming, for simplicity, that each technician must do ten windshields a week or be fired. Reality
would be more complex; a hardworking technician is likely to be allowed an occasional week where he
does less than ten, as long as it doesn’t happen too often and he otherwise has a good performance
record.
‡
RE/MAX offers prospective agents options on this basic scheme, options that offer a measure of
insurance and that, therefore, probably dull the screening effect to some extent. See
https://sites.google.com/site/joinrhe/compensation-plans.
§
The use of the term “option” is a bit imprecise. While it is true that the employee can choose within
some limits when and whether to exercise his rights to buy shares, in financial markets, the term
“option” is generally used to describe a security in zero net supply. When and if the employee exercises
one of these options-cum-compensation, the company creates new shares of common stock, diluting the
value of existing shares. The proper financial term for this sort of financial instrument is warrants.
¶
This argument assumes the individual can freely purchase shares in the market. If he has inside
information that the company is worth more than its current market price, he might be barred from
purchasing shares by insider-trading rules but allowed to exercise options he holds. In which case, he
might exercise options even when the market price is below the strike price.
#
Explanation 2 can still hold, albeit with reduced force, if FASB’s rule for how stock options must be
expensed combined with the value the employee puts on the options relative to cash makes the granting
of options a cheaper alternative to paying cash.
3
Pay for performance, in the form of piece-rate pay or year-end bonuses or,
more generally, any rewards for performance, can be a powerful
motivational tool. That is clearly shown in the case of Safelite. But, as the
other anecdotes from the last chapter make clear, for jobs more complex than
that of Safelite technician, pay for performance can be difficult to get right;
and it may even be dysfunctional.
You, presumably, are interested in answers to the following questions:
US/Canada 30.0%
Europe 28.0%
Age
Organizational rank
COO 4.3%
VP/partner 19.3%
Sex
Male 86.0%
Female 14.0%
Functional specialty
Finance 9.2%
Accounting 0.5%
Marketing 7.2%
Ops./prod./mfg. 6.3%
Other 15.5%
Industry
Manufacturing/construction 14.5%
Marketing/retail 6.3%
Consulting/advisory/education 5.3%
Other 19.3%
Which of these five channels is MOST descriptive of what motivates your own best work?
Which is MOST descriptive of what motivates the best work of your direct reports?
When it comes to motivating their own best work, 27.5% of the respondents
said contributing to success of the organization and 25.6% said interesting
and exciting work is most descriptive, while only 15.4% said tangible
personal rewards. For their direct reports, 26.6% said exciting and
interesting work, 23.6% said contributing to organizational success, and only
17.4% said tangible personal rewards.
Forcing the survey respondents to pick one most descriptive motivational
channel doesn’t necessarily give a full picture of their opinions about the
effectiveness of each. So I also asked them to rate each of the five
motivational channels on a six-point scale: not at all effective; of limited
effectiveness; effective but not very effective; very effective; extremely
effective; and only this channel is effective. The full results are shown in
Table 2, where I also compute average or mean scores (where not at all
effective = 1, etc.), and the standard deviations of the scores. The full results
show that, although tangible personal rewards—the stuff of pay for
performance—are not ineffective, they lag significantly compared to
interesting and exciting work, contributing to organizational success, and
even intangible personal rewards.
The participants in SEP average 47 years old, and the median participant
in terms of rank is a senior VP/partner. Would we get similar results by
surveying younger and less senior managers? I’ve conducted a very similar
survey with students in the Stanford GSB’s MBA program; the differences
were that the MBA students were asked to discuss motivation on their last
job before enrolling, and instead of their direct reports they were asked to
discuss their peers at their last job. My survey was spread across two years
of students who were enrolled in a course on HRM that I taught in the spring
quarters of 2014 and 2015. I received 240 responses, representing an 80%
response rate, so this is fairly representative of the MBA population (or, at
least, that segment that chose to take a course in HRM). The demographics
for the respondents are in Table 3, and their responses to the five-channel
questions are in Table 4.
Table 2. Responses from SEP Participants on the Five Motivational
Channels
Intangible Contributes
Tangible Interesting Contributes to
personal to
rewards; & exciting greater social
rewards; organizational
e.g., pay work purpose
e.g., praise success
Intangible Contributes
Tangible Interesting Contributes to
personal to
rewards; & exciting greater social
rewards; organizational
e.g., pay work purpose
e.g., praise success
Not at all
0.5% 0.0% 0.0% 0.0% 1.0%
effective
Limited
6.3% 3.9% 0.5% 1.0% 10.6%
effectiveness
Extremely
27.5% 32.4% 58.0% 52.7% 24.6%
effective
Only this is
1.0% 4.3% 7.2% 7.2% 3.4%
effective
Standard
0.95 0.71 0.76 1.07 0.90
deviation
Not at all
0.0% 0.0% 0.0% 0.0% 0.5%
effective
Limited
8.7% 1.0% 1.9% 2.9% 20.3%
effectiveness
Extremely
21.3% 27.5% 51.2% 30.9% 13.0%
effective
Only this is
1.0% 1.0% 4.3% 1.4% 0.5%
effective
Standard
0.91 0.75 0.74 0.83 0.97
deviation
US/Canada 70.8%
Europe 8.3%
Age
Sex
Male 56.7%
Female 43.3%
College major
Economics 20.8%
Business 16.7%
Engineering 21.3%
Humanities 14.2%
Functional specialty
Finance 23.8%
Accounting 0.4%
Marketing 8.3%
Ops./prod./mfg. 5.4%
Other 13.8%
Industry
Manufacturing/construction 4.2%
Marketing/retail 5.0%
Consulting 20.4%
Education 4.2%
Entertainment 3.8%
Other 8.3%
Table 4. MBA-Student Responses to the Five-Motivational-Channel
Questions
Intangible Contributes
Tangible Interesting Contributes to
personal to
rewards; & exciting greater social
rewards; organizational
e.g., pay work purpose
e.g., praise success
Intangible Contributes
Tangible Interesting Contributes to
personal to
rewards; & exciting greater social
rewards; organizational
e.g., pay work purpose
e.g., praise success
Not at all
0.4% 0.0% 0.4% 2.1% 5.0%
effective
Limited
7.5% 2.9% 0.8% 15.4% 13.8%
effectiveness
Extremely
26.3% 52.5% 65.8% 20.8% 22.9%
effective
Only this is
1.3% 2.9% 8.3% 1.3% 3.3%
effective
Standard
0.94 0.81 0.76 1.08 1.21
deviation
Not at all
0.4% 0.0% 0.0% 2.5% 5.0%
effective
Limited
4.6% 1.7% 1.3% 19.6% 26.3%
effectiveness
Extremely
33.8% 40.4% 49.2% 14.6% 10.4%
effective
Only this is
1.7% 0.4% 4.2% 0.0% 0.4%
effective
Standard
0.95 0.74 0.83 1.02 1.10
deviation
Given the back half of the previous chapter, the second conclusion
shouldn’t come as much of a surprise, given all the complications and
limiting factors of pay for performance. But how do we understand what
does work? And, insofar as there is no single answer—insofar as effective
motivation must be tailored to the specific work context—how can we think
in systematic fashion about the connection from context to effective
motivation?
To answer this last question, and in particular to understand what might
be effective in the context of your organization, you first must understand the
employment relationship both as an economic and as a social relationship.
If you’ve studied economics, the picture you are likely to have in your
mind of “economic relationships” is the hoary economic model of supply
equals demand. The economic theory that underpins pay for performance, in
essence, is based on supply equals demand. It conceives of employment as a
discrete and one-time transaction between employer and employee. The
employer tells the employee, “This is how your compensation will be
determined.” And the employee then chooses whether to take the job at all
and, if so, how hard to work and on what tasks.
In some contexts, that model of employment works well. But in cases
where employment is not a one-time transaction but an enduring relationship
between employer and employee, we must rethink the nature of the
“transaction,” both in terms of the economic and the social exchange. So,
fasten your seatbelts: For the next two chapters, we explore employment, and
specifically employment relationships, in both economic and social terms.
_____________
*
One thing that makes research of this sort difficult to trust has the technical name of endogeneity.
Firms presumably choose their motivation schemes based on their specific context (strategy) and
environment. Researchers can say that they are “controlling for observable differences” among firms in
their sample, but the adjective “observable” in that phrase is there for a reason.
†
The appendix will supply measures of statistical significance of these differences.
4
Because Bob has these assets at risk—because it will be costly for him to
walk away from his job—he is the victim of a holdup (which, believe it or
not, is the technical term used by economists to describe this situation).
Management at Zephyr knew that Bob, like other employees, faced these
costs of quitting—the technical term is transaction costs—and they knew
that they could, in consequence, take advantage of him to some extent,
keeping him at a posting where he was valuable to the company, even if it
was not the posting he wanted.
But understand that while Bob has assets at risk, so does Zephyr. Bob has
gained experience at Zephyr and, even more significantly, he has become
knowledgeable about the project on which he is working. Therefore,
Zephyr’s next best alternative to dealing with Bob is not as good as keeping
Bob on the job. If the gap in values to Zephyr between keeping Bob in
Chennai and their next best alternative is wide enough, maybe Bob can turn
things around and hold up Zephyr by telling them, “If I’m so valuable to you
on this project, give me an immediate bonus. If you don’t, I’ll quit.” If Zephyr
believes Bob will quit without this bonus, and if the size of the bonus is less
than the gap for Zephyr between having Bob and not, Zephyr might well
decide that it is prudent to pay.*
This is why supply equals demand isn’t the right economic model. What
drives supply equals demand is that every market participant has a nearly-as-
good alternative; namely, to transact with another buyer or seller at virtually
the same price, for virtually the same good. In a market where this condition
is met, neither buyer nor seller can hold up the other side. Neither side can
say “For you, I’m the uniquely best trading partner you have—better by a
nontrivial amount than your next best alternative—and I will use that fact to
extract some extra value (money) from you.”
Contrast Bob’s situation with my relationship with my grocer: For many
years, I bought most of my groceries at a local family owned and operated
supermarket. The relationship that I had with the owners of this grocery was
open-ended, just like Bob’s relationship with Zephyr: I didn’t know what I
would buy six days, let alone six months, hence, and they gave me no
guarantees what prices they would be charging in the future, or whether they
would continue to carry the particular cuts of meat that I would want. And if
they kept an inventory of those particular cuts of meat, they had no assurance
that I’d show up to buy. But if, one day, I decided that the relationship was no
longer working for me, the cost of transferring my business to the local
Safeway would be trivial. And if I failed to show up, there would be plenty
of other customers for them to serve, some of who would be happy to buy
what I did not. Hence, supply equals demand, applied day by day, works as
an economic model in this case.
And supply equals demand would be an adequate model if, from the
outset, Bob and Zephyr could put in place contractual guarantees for What
next? over the duration of his employment. At the moment he was comparing
jobs, he didn’t have assets at risk; if he knew (and could guarantee) exactly
what would transpire for as long as he worked for Zephyr, he could (at least,
according to economic theory) compare the Zephyr job offer with others and
make a free choice. If Zephyr could lock in exactly what Bob would do, they
could decide whether to make him an offer and on what terms.
It is the combination of the incomplete or open-ended nature of the
relationship and the fact that, once the relationship is consummated,
parties have assets at risk, that takes us away from supply equals demand.
If supply equals demand is not the right economic model, what is? For
enduring employment relationships, economists have come to rely instead on
a model based on three interrelated concepts: governance, credibility, and
reputation.
Governance
Governance refers to the rules that govern how the relationship evolves;
essentially, it comes down to which parties are allowed—either formally or
informally—to make various decisions about, What next?
In some cases, called hierarchical governance, one of the parties
involved in the transaction makes most of the What-next? decisions. For
instance, and in a very different context, Toyota outsources to subcontractors
the manufacture of many major subassemblies that go into its cars, on the
basis of ongoing relationships with those subcontractors. The formal contract
between Toyota and each subcontractor is very short and succinct; in
particular, the contract specifies neither the quantities Toyota will buy nor the
prices Toyota will pay past the first few months of the contract. Instead,
Toyota, by convention that is completely understood by both parties to the
relationship, makes these decisions on a take-it-or-leave-it basis. A
subcontractor reserves the right to quit at any time. But, having invested in
their relationship with Toyota, subcontractors rarely quit. This, of course,
makes them very susceptible to a holdup. We’ll explain what protects the
subcontractors—and moreover gives them the motivation to invest in their
relationship with Toyota—shortly.1 Traditionally, employment relationships
are similarly governed: The employer tells the employee what he will do on
a day-to-day basis; the employee’s options are to do what he is told or quit.
In other cases, decision-making authority is shared. Sometimes, the rule
is that there is a status quo, and it takes consent by both parties to change the
status quo. Or, if no agreement is reached, the relationship simply dissolves.
In other cases, different parties have decision rights over different specific
decisions; for instance, one party (the seller, say) decides on the price to be
paid; the second party (the buyer) decides on how much to buy at that price.
In the context of employment, this sort of bilateral governance is observed
in particular when employees are represented by a union.
And in still other cases, decision-making authority is vested in some third
and supposedly neutral party. An example is a labor agreement between a
corporation and a labor union in which disputes are settled by binding
arbitration by a panel of neutral labor experts. In some cases, the arbitrators
are free to propose a compromise. In other cases, the rules of arbitration
state that the two principal parties must state their alternative positions, and
the arbitrators are constrained to pick one or the other of these. Governance
in which a third and supposedly neutral party is part of the picture is called
trilateral.
These three forms of governance are ideal or pure forms. In real-life
examples, you nearly always have a mix. For instance, even in the most
hierarchical employment situation, where the employer tells the employee
what to do, the employee retains the right to quit.
How are these decision rights determined? In some cases, it is a matter
of law. To give an obvious example: Laws against slavery and indenture
provide all employees with the right to quit at will (although noncompete
clauses in employment contracts sometimes limit what the employee can do
after quitting). Contract law, and labor law in particular, provide further
limitations on how decision rights are assigned; for instance, an employer
cannot compel an employee to take illegal actions or fire an employee who
refuses to do so.
In other cases, decision rights are assigned contractually, as in “take-or-
pay” clauses between a vendor and its customers. In employment contexts,
decision rights can be the subject of formal negotiation by the parties
involved, especially when employees are organized into a union. (And, when
unions are involved, there are legal restrictions on the agreements that labor
and management can reach.)
But perhaps most often, decision rights are a matter of mutual
understanding: Subcontractors to Toyota understand the way Toyota does
business, namely Toyota calls nearly all the shots. This is never explicitly
stated, and it certainly isn’t part of the formal contract between Toyota and
the subcontractor, which instead is an aspirational paean to cooperation,
mutual agreement, and mutual benefit. But a company that wants to work with
Toyota knows, if only through the grapevine, that it is Toyota’s way or the
highway. Or to return to Bob and Zephyr, Bob was told up front that he’d be
given his first few assignments without discussion; when he accepted the job,
he implicitly accepted that this was “how things are done at Zephyr.”
A different but related question is, Why are decision rights assigned as they
are? Why design a take-or-pay contract? Why does Toyota insist that it call
the shots in its relationship with its subcontractors?
When it comes to legal restrictions on governance, we get a mix of “what
lawmakers think is socially good” with interest-group politics. To give an
example, Tesla, the automobile manufacturer, is legally barred from selling
its cars in the state of Michigan. Instead, if it wants to sell cars in Michigan,
it must set up an independent dealership network. The laws in this case were
created long ago to protect the interests of independent dealers (and other
franchise operations) from being held up by their upstream “partners”; today,
the laws are maintained well beyond their original purpose by organized
groups—in this case, dealership organizations in Michigan—because the
laws benefit the group’s members economically.
But beyond legal restrictions, governance is a matter of design and
choice. In contracts, the parties negotiate over who decides what when it
comes to What next?
In cases like Toyota or Zephyr, one party says, “This is what works for
me, and if you want to deal with me, you should understand and agree.” Of
course, negotiating power and the desire to be in control play their parts
here; Toyota gains from calling nearly all the shots, and prospective
suppliers find it so desirable to form a relationship with Toyota that they
agree to this arrangement. But, more than this, Toyota insists on this
arrangement because it is efficient. Any assignment of decision rights has
implications for how smoothly and well What next? will be determined and
then implemented. Making good What-next? decisions and implementing
them smoothly means a more valuable transaction overall, and more value
means more for the parties to split between them. So, in particular, as you are
thinking about what sort of governance provisions you want in your
organization, you should be asking, What’s the best, most efficient way to get
to good decisions?
Start with information and ability: Which party is better able to
formulate the “best” decision and which party has the best information
pertinent to this decision? One reason for Toyota to call nearly all the shots
is that Toyota is in the center of a large array of bilateral relationships, and it
knows best about coordinating the activities of its many suppliers. Similarly,
the classic model of employment—where the employer directs the activities
of employees—is efficient because the employer specializes in
understanding how to achieve coordination among her many employees.
Note, however, that when it comes to Type-K employees, the employer
no longer knows what is best to do; it becomes efficient for the employee to
choose what he does and how he does it. For instance, when I was associate
dean of Stanford GSB, it made sense for me to tell faculty members which
courses they would teach, in which quarters, on which days, and at which
hours, because I had the job of coordinating the different course offerings
available to students. (Of course, I asked faculty members for their
preferences, which I tried to accommodate. But the final say belongs to the
dean.) But it made no sense for me to tell my colleagues how to teach the
courses to which they were assigned.
In many cases, neither party to a transaction has a monopoly on valuable
information relevant to some particular decision. One party knows some
things that are important; the second party knows other things, and the best
decision emerges if they pool their information.
Toyota understands this and still it wants to call all the shots. So, in the
day-to-day management of its relationships with its suppliers, Toyota spends
a lot of time and money understanding in detail the inner workings of those
suppliers. As dean, I always asked my faculty colleagues about their
preferences and their constraints concerning what and when they would
teach, and I tried, but only sometimes succeeded, in giving them what they
requested. As employer, if there are decisions you want to be able to make
concerning your employees’ activities that are best made based on
information that is naturally held by them, then it is your job—a crucial job—
to spend time and effort learning enough of what they know, so you can make
a good decision.
Credibility
If information and ability are crucial considerations when it comes to
efficient governance, credibility is just as important. For an employment
relationship to work well, decisions should be made by the party that can be
trusted not to exploit their decision rights.
This takes us back to holdups and the assets at risk for each party. Zephyr
might retain most decision rights about where Bob will be posted and for
how long, because they are coordinating many different employees at many
different sites. But, as already discussed, this leaves Bob open to a holdup.
When, at the outset, the Zephyr recruiter told Bob about the job, she told him,
“For a while, at least, you will be told what to do, where, and with whom.”
Bob was under no illusions about who would have these decision rights. So
an important consideration for Bob was, “If I let Zephyr management control
my work life to that extent, how much can I trust them not to take advantage of
my situation?”
The issue of credibility becomes even more important when the decision-
making party wants the other party to make investments in the relationship.
Some of the assets that Bob has at risk in his relationship with Zephyr are
created naturally, such as friendships with co-workers and loss of income
and risk of a bad recommendation if he quits. But beyond these assets, he can
choose whether to create more. Suppose, for instance, Alice asks Bob to
attend evening classes, to learn Tamil, so he can better communicate with the
local workers on the job. Bob knows that he can refuse this request; it goes
beyond what he is expected to do, and it is his choice. Taking these courses
will cost Bob some of his free time, and while speaking rudimentary Tamil
might be an asset he can employ in another job, the odds are that it will not
be that valuable to him except on this job or some other job in Southern
India. If he suspects that Zephyr is going to use their decision rights to hold
him up, he might think that he will get little return from an investment he
would make in learning Tamil. Put the other way around, if Zephyr (and
Alice) wants Bob to make this personal investment, they need to be credible
in saying, “This will be good for us, and we’ll make sure in the future that
you are rewarded for undertaking this investment.”
So what makes one party more credible? Formal contractual guarantees,
including guaranteed appeals processes, help, of course. But, in this context,
the third concept, reputation, often plays a decisive role.
But, before going there, three grace notes are worth mentioning.
Employee Voice
A standard “principle” in business strategy is that your business will be
better off the less powerful are your suppliers and customers. (This principle
is enshrined, in particular, in the famous “Five Forces” of Michael Porter.2)
But consider how Toyota treats its frontline suppliers: Toyota arranges for
those suppliers to meet together at regular intervals; it provides the venue
and pays for expenses incurred. Toyota does this for several reasons: First,
Toyota wants its suppliers to educate each other, sharing best practices and,
in particular, best practices in dealing with Toyota. If a new supplier is told
by Toyota, “This is how we do things,” it helps if the new supplier hears the
same message (that is, “This is how you do things if you supply Toyota”)
from a long-time, successful supplier to Toyota. Second, Toyota can use this
sort of forum to fine-tune its reputation with its suppliers, of being “tough, but
fair.” Toyota will, on occasion, insist on something with a specific supplier
that may seem, to that supplier, more tough than fair. An aggrieved supplier
can discuss the specifics at one of these meetings, with the other suppliers
acting as a jury of peers: Typically, the peers will render a verdict of
“Toyota is not guilty.” And on the rare occasion that the verdict is “Toyota
probably went too far,” Toyota will know to back off.
A third reason for promoting this sort of meeting is that it makes Toyota’s
“promise” not to abuse individual suppliers more credible. It does this by
empowering the suppliers collectively. If Toyota were to act abusively in its
relationship with a single supplier, that supplier could quit the relationship,
which would hurt Toyota a bit. But, because of these meetings, such abusive
actions by Toyota would be noticed by all the suppliers at the forum. It
would cause them (all the suppliers) to worry, “How will Toyota treat me?”
It might incline those suppliers to look for contractual guarantees with
Toyota, which would reduce Toyota’s cherished and valuable flexibility. It
might incline them against making sunk-cost investments in their relationship
with Toyota, in fear of being held up by Toyota. In other words, Toyota, by
promoting this sort of supplier forum, turns its reputation with each supplier
into a reputation with all its suppliers. Toyota, in each individual, bilateral
relationship, has its general reputation more at risk. Hence Toyota is more
careful in its bilateral relationships; it has enhanced credibility in each of
those relationships.
This is not always a good thing for Toyota. In some cases, Toyota might
want to make an exception to its general policies with an individual supplier,
an exception that would benefit both Toyota and the specific supplier. If it
could do this “secretly,” without its other suppliers knowing, it might do so.
But since word of the exception is bound to come out, Toyota has to be
careful not to make such exceptions, to preserve its reputation for doing
business in a certain way.
What is the connection with the principle that you want weak suppliers
and weak customers? If your relationship with your suppliers or your
customers is based on your reputation, what keeps you “honest” or credible
is the threat that suppliers or customers will hurt you if you don’t act
according to that reputation. If they can’t hurt you (enough), they can’t trust
you. So, weakness per se is a disadvantage, if the nature of your relationship
is one based on reputation-derived credibility.
In the context of employment relationships, this takes us to the issue of
worker collectives such as labor unions. Ask the average American manager
about dealing with an organized workforce, and the answer will, in most
cases, be “Avoid that if at all possible.” But empirical analysis of unionized
versus non-unionized settings paints a more nuanced picture. For a variety of
reasons, including the promotion of trust and more efficient assignment of
decision rights, having an organized workforce may promote efficiency in the
relationship between labor and management. In essence, it gives labor an
efficiency-promoting voice in labor-management relations. But having an
organized workforce also strengthens the bargaining power of labor. These
two effects, on average, net out a bottom line for management that is worse
with a union than without; labor uses its bargaining power to take a slice of
the “pie” so large that management’s smaller share of the larger pie with a
union is less than its larger share of the smaller pie without. ‡ But in cases
where labor and management have a positive and constructive relationship,
the enhanced efficiency can overcome the bargaining power; the enhanced
size of the pie allows both parties to have more pie in absolute terms. The
moral here is: As employer, your bottom line might improve with an
organized labor force, if you take the time and trouble to promote a
constructive relationship with the unions with which you must deal.§
2. But the world of supply equals demand is based on (at least) three
important conditions: (a) there must be a lot of suppliers and a lot
of demanders, so competitive forces can work; (b) the thing—in
this case, the service—being bought and sold should be a
commodity; (c) the market must be made; someone or some
institution must bring buyers and sellers together.
These conditions are why I twice inserted the phrase “if the gig
economy functions well” in my first point: (a) a tile setter who
works as a subcontractor for one general contractor is a lot closer
to employee-employer, and further from supply equals demand,
than the computer programmer who puts her services up for sale at
Upwork; (b) there are quality dimensions to how well the job is
done, and so some mechanism for transmitting reputations must be
present (moreover, the nature of the job must be relatively clear
from the start; “very Type-K” jobs are not good candidates for
gigging); (c) an intermediary that makes the matches—and also
helps with reputation transmission—is important. (Uber, for
instance, is both an efficient intermediary—it has the information
on both sides of the market needed to make efficient matches—and
a trustworthy intermediary—it collects reviews on both clients and
drivers. An Uber driver who accumulates a few bad reviews
quickly becomes an ex-Uber-driver.)
The answers to “Who do you aspire to be?” are similarly dimensioned. And
for “To what extent do you fall short of those aspirations?,” answers can
range from, “I’m just where I want to be” (along a specific dimension) to “I
fall short in the following ways . . .” And, of course, individuals can and will
give compound answers: “I’m an ambitious blue-collar worker at Sun
Hydraulics. In my spare time, I coach youth soccer, and I’m quite good at
dealing with young boys who are just learning the game. But, most
importantly, I’m the mother of my children and the wife of my husband:
Family is everything to me.”
Our self-perceptions are important because we try to live up to our
perceptions of ourselves or, where we are less than satisfied with those
perceptions, to our aspirations. We desire to exercise and exhibit
competencies of which we are proud. We strive to behave in ways that, we
believe, members of social groups to which we belong should act. We
associate ourselves with groups whose actions conform to our values and
disassociate from groups whose actions conflict with those values.
And, a feedback loop can exist here: When we behave in a manner that is
inconsistent with how we perceive ourselves, we can dismiss what we did
(“I made a one-time mistake”), we can invent situational excuses (“I did that
because . . .”) or, especially when we consistently behave in this fashion, we
can reframe our self-perception.
Different aspects of one’s self-perception can, in specific contexts, imply
different behaviors. Imagine that the ambitious Sun Hydraulics employee,
youth soccer coach, and mom must choose how to allocate her time between
staying overtime to finish a rush order, making soccer practice with her team,
and staying at home to care for a sick child. Of course, her choice depends on
its consequences; if her assistant coach is fully capable of running this one
practice, and if she can enlist another family member to care for the sick
child, then perhaps she will stay and work overtime. But the importance of
each aspect to her self-perception is also crucial. If “family is everything” to
her, caring for the sick child is likely to be the choice she makes.
This probably seems obvious. But what may be less obvious is that self-
perceptions, especially about social identity, can be primed: An individual’s
behavior can be significantly affected if, prior to the behavior taking place,
the individual’s connection to one group or another is emphasized to the
individual. In the one study,1 the authors primed a group of Asian women
students prior to the students taking a math test: One-third were primed with
material that emphasized their Asian identity; one-third were primed with
material that emphasized their sex; and one-third were primed with material
that was race and sex neutral. Conforming to stereotypes that Asians are good
at math and women are not, the first third performed (on average) best on the
subsequent test and the second third performed worst, with statistically
significant differences. In a second study,2 students were primed in one
treatment to think of themselves as budding scholars and the other treatment
as college “socialites.” They were then asked their preferences between
pairs of objects where, in each pair, one object had more appeal to a scholar
and the second to a socialite. The choice behavior was affected in the
obvious fashion and, moreover, the postchoice satisfaction of a subject could
be decreased by re-priming the subject in the “other” direction.
Of course, an employee’s work, the people with whom he works, and the
organization for which he works all play an important role in the individual’s
perceptions of self. Put the other way around, the nature of the employee’s
connection to his job—where “job” here is meant to combine the work he
does, his co-workers, and the organization for which he works—is
influenced by his general perception of himself and the role his job plays in
that self-perception.
A useful distinction in this respect is between instrumental connections
to the job—where the job is a means toward some desired end—and
expressive connections, where aspects of the job provide primitive value to
the employee. The ultimate in an instrumental connection is expressed by the
employee who says “I work for the money I’m paid. Pay me more than I’m
currently making, and I’ll quit my current job and work for you.” Or consider
Bob: To the extent that he is working with Zephyr to build his résumé, make
some money, and see parts of the world he hasn’t seen, all with the aim of
someday going back to school and, say, getting an MBA, then his connection
to the job is largely instrumental.
On the other end of the spectrum are expressive connections to:
■ The work the employee does. This can take several forms. Bob
could take pleasure in exercising his skills as a civil engineer,
tackling difficult construction projects. He might enjoy building on
those skills. (Of course, if he wants to build his skills so he will
be promoted or otherwise moved to a better job, this is more
instrumental, as would be a desire to exercise his skills to gain the
esteem of co-workers.) Or the work can be a “mission” for the
employee; think of a pharmaceutical researcher, looking for drugs
to help treat a particular form of cancer, or of a doctor working
for Doctors Without Borders, or of employees at Artisans’
Alliance, working to improve third-world economies.
average for
average for average for average for
direct
self self peers
reports
Results are provided in Table 5. For each group, the average or mean
score given to the motivator is provided, which respondents were asked to
assess on the following scale: no impact = 1; negligible impact = 2; limited
impact = 3; moderate impact = 4; substantial impact = 5; powerful impact =
6; I [my direct reports/my peers] live and die according to this motivator = 7.
These data tell a number of interesting stories. Suppose we put the eight
options into two buckets: four categories of motivator that are extrinsic
(supplied by others)—benefits, job security, pay, and praise—and four that
are intrinsic (related to inner feelings of satisfaction)—feeling good, learning
and growing, acquiring and practicing skills, and doing worthwhile things.
These buckets reveal two key observations. First, for the SEP participants,
each of the intrinsics outaverages each of the extrinsics for themselves, and
for the MBA students, only praise outscores some of the intrinsics. Second,
for the SEP participants, the average score for “self” for each of the
extrinsics is less than the score they give their direct reports. For the
intrinsics, the reverse is true. The same pattern holds for the MBA students,
with the exception of praise.‡
Since the respondents give different (relative) answers to the impacts of
the different categories on themselves versus their “others,” the obvious
question to ask is, Are the respondents more accurate when assessing the
impact on themselves or on their direct reports or peers? If the assessments
are more accurate for direct reports and peers, then the survey results suggest
that pay and praise are most impactful. If they are more accurate for
themselves, then the intrinsic categories have more impact, overall.
So which is it? Why do the respondents give such different results for
themselves than for their direct reports or peers? It could be that both sets of
assessments are accurate: The respondents are, after all, a self-selected
group; the SEP participants chose to spend six weeks at an executive
education program; the MBA students chose to leave work to go back to
school. Maybe, on such grounds, the respondents are different from their
direct reports or peers. Or perhaps the respondents are deluding themselves.
After all, most people would prefer a self-image that emphasizes a “noble”
desire to do good things and to learn and grow, rather than a self-image of
being driven by “craven” desire for money, recognition, and so forth. A third
possibility is that the respondents are mistaken about their direct reports and
peers. Perhaps their direct reports and peers have similar motivations as do
they, and they just assumed (or focused on) “the worst.”
We have no way to know which of these are true and, in fact, I believe
there is some truth to all three. But I believe that most managers
underestimate the extent to which their employees and peers (and bosses) are
motivated by the more intrinsic categories of motivator. So, as you look at the
long list of “things that motivate people,” don’t be quick to think that money
is everything or that it dominates the other items on the list.
Having said that, I caution you not to overprocess the data in Table 5,
which present average ratings for the eight categories from fairly large
populations. The answers provided by these populations exhibit a great deal
of dispersion. I give some specific indications of this in the appendix, but
suffice it to say that, within my samples of SEP participants and MBA
students are substantial minorities for whom, based on their responses,
money is indeed most important.
One reason why employers tend to overestimate the importance of pay is
that they underestimate how much the “process” of work matters to
employees; they tend to think of work as instrumental for their employees
rather than expressive. The list should remind you that process matters. No
better example of this can be offered than the general social norm of
reciprocity.
■ Jill tells Bob that Zephyr wants to make amends, and asks Bob to
suggest a one-time bonus amount that he thinks is fair
compensation. Bob asks for $15,000 and, after some negotiation,
Jill and Bob agree on $10,000.
Self-Perception Theory
Imagine that a co-worker of Bob—call him Bill—is failing in an assignment
given to him (to Bill) by Alice. Imagine that Bob, who thinks Bill is an “okay
guy” and who has the skills needed to assist Bill, works overtime to help Bill
succeed, so that (in the end) Bill does succeed. Imagine that this cost Bob a
lot of his free time.
Self-perception theory holds that, insofar as Bob instinctively lent a hand
to Bill, Bob will apply attribution theory to himself, asking himself ex post,
“Why did I do that?” Several explanations are possible:
And so forth. The point, according to the theory, is that, whatever explanation
Bob offers to himself, this becomes a stronger piece of perception of himself,
affecting his later behavior. If, say, he settles on the second explanation, then
he is more likely later to go out of his way to help his (now perceived to be)
very good friend Bill. If he settles on the fourth explanation, he is more likely
to look for opportunities to impress Alice, to generate a bigger bonus.
We’ll see in the next chapter how you, as an employer, can use self-
perception to “mold” your employees in ways that improve their
performance and how, if you are not careful in what motivational techniques
you use, self-perception can mold your employees in adverse ways.
_____________
*
In fact, in these sorts of economic models, it is worse than this suggests. Bob’s evaluation of the
benefits of staying with Zephyr involves the results of decisions he will be called upon to take later, so as
he evaluates the current stay/leave decision, he must contingently analyze all the future decisions he will
face if he remains and all the future decisions he will face if he leaves.
†
The basic idea of this survey, and in particular the eight specific categories of motivator, come from
Chip Heath, “Lay Theories of Motivation Overemphasize Extrinsic Incentives,” Organizational
Behavior and Human Decision Processes 78 (1999), 25–62. In Heath’s original study, he asked
respondents to rank order the eight categories in terms of impact, rather than rating them on a given
scale. But the conclusions he reaches are virtually identical to those I draw from my version of his
work.
‡
Are these differences statistically significant? For the most part, they are; in many cases they are
remarkably statistically significant. The appendix provides critical p-values for the differences.
§
Obviously, there are tax implications to consider as well. But hold those aside.
¶
I asked several colleagues who are trained in social psychology: Could this be explained as an example
of priming, in that the employee is being cued, perhaps nonverbally, to regard himself as a member of a
trusted (or untrusted) group? Could it be viewed as an example of gift exchange, in which the employer
offers the gift of trust, and the employee reciprocates by being trustworthy? My informal survey
gathered all four patterns of yes/no responses, a testament to the quotation from Dale Miller offered
earlier.
6
Also, turnover within the nursing staff was reduced—at a time where many
hospitals had nursing vacancies, Beth Israel maintained a list of nurses from
other hospitals seeking a job at BI—and standard job-satisfaction metrics for
the nursing staff increased significantly. Rabkin had decided to implement
primary nursing to take advantage of the information held by nurses about
patients. He achieved this aim and, at the same time, motivated the senior
RNs at Beth Israel to provide truly consummate effort, entirely without any
form of incentive pay and in a manner that left them more satisfied with their
jobs.
It is difficult, to say the least, to explain this motivational effect with the
economic theory of incentives. Nursing is very much a Type-K job, and the
economic theory of incentives suggests that, due to the multitasking nature of
the nursing function, incentives for primary nurses, who must balance a
variety of tasks, will be difficult. One can even make a good case on
economic grounds that functional nursing, where individual nurses are given
jobs with fewer and more consistent tasks, is the way to go. So, to explain
why primary nursing worked so well in terms of motivation, we turn to
theories of motivation grounded in cognitive and social psychology, to see if
we can find explanations there.
We begin by reiterating that psychology, in comparison with economics
and its one theory of incentives, provides a number of theories of motivation.
This chapter provides brief accounts of some of the more prominent and
useful of these theories. The list of theories of motivation provided here is
not exhaustive; social psychology provides others. And, importantly, these
theories are not exclusive; more than one can be applied to a particular
situation. (As we’ll see when it comes to Beth Israel and primary nursing, at
least two of the theories I’ll describe help us to understand why,
motivationally, it was such a success.)
Often when more than one of these theories apply, they are
complementary; they differ because they are based on different accounts of
how and why people behave as they do and on different “psychological
levers” you can employ. But they lead in the same direction; the prescriptions
that one theory suggests will strengthen the impact of the prescriptions
suggested by the other theories. (This is so, for instance, in the case of Beth
Israel.) Sometimes they are simply orthogonal. Occasionally, they offer
conflicting accounts and, more importantly, conflicting prescriptions.
This multiplicity of theories—further increased when you also fold in the
economic theories drawn from Chapters 2 and 4—means that you, as
practitioner, have the difficult job of sorting among them. That won’t be easy,
but (as I tell my MBA students when faced with a multiplicity of frameworks
that might be applied to a specific situation) that’s why you get the big bucks.
Expectancy Theory
In expectancy theory, the employee consciously considers three dimensions
of every set of actions she might take:
The theory predicts that an employee will choose actions that maximize her
chances of receiving rewards that she values the most. Some versions of the
theory make this prediction exact, by asserting that she computes the product
of the two probabilities—the probability of achieving the results management
desires times the probability that management will reward her for doing so—
times a measure of her value for the prize. But, we don’t need anything so
formal or rigid. The theory asserts, essentially, that she takes actions that she
believes are more likely to lead to her perceptions of management’s desires,
that she believes are more likely to be rewarded, and that provide rewards
that are more valuable to her.
What are the managerial implications?
Goal-setting Theory
According to goal-setting theory, employees are motivated to achieve goals
that are set for them (by management or by themselves); this is a version of
expectancy theory, where the reward is achievement of a goal itself.
Goal-setting theory asserts that, to be effective, the goal should be
■ Time-bound, meaning that the employee can see from the start the
proverbial “light at the end of the tunnel.”
The literature refers to these conditions with the acronym SMART (Specific,
Measurable, Achievable, Relevant, Time-bound): Goal-setting is an effective
motivational tool when (and only when) the goal is SMART.
In addition, goals should be somewhat challenging: Achievability is part
of SMART. But, on the other end of the spectrum, goals that require little if
any effort to achieve provide no sense of accomplishment or satisfaction
when achieved, and so provide little motivation.
While setting SMART and yet challenging goals may provide motivation,
for Type-K jobs, many of the problems that arise in pay-for-performance
schemes recur: When the employee’s job involves several distinct tasks, do
you set a single goal that encompasses all the tasks or a set of goals, one for
each task? And, whichever you do, how does the employee allocate her time
among them? Suppose you set different goals for different tasks. Since the
reward is meant to be a sense of accomplishment or fulfillment, will the
employee spend most of her time attending to those goals that give her the
best chance of quick success? Or will the employee be more motivated to
avoid failure on any of the goals? And what if achieving success in some
tasks is intrinsically more satisfying for the employee than success in others?
So perhaps a single goal that blends measurable accomplishments on
different tasks is better. But, as in the discussion of multitasking in pay for
performance, what if results in some tasks are easily measured while others
are difficult to measure? And, since success in Type-K jobs often involves
creativity, how do you simultaneously set goals that are specific, when the
tasks to be done are ambiguous ex ante.
Imagine a job that incorporates two distinct types of tasks. In one type of
task, the organization wants the employee to strive for brilliant success, even
at the risk of failure, because the organization doesn’t suffer much from the
failures but feasts when an individual achieves great success. Such tasks are
called star tasks. Think, for instance, of a researcher working for a
pharmaceutical firm: Failing to find a new drug doesn’t cost the organization
very much in the grand scheme of things, but one breakthrough drug can mean
huge profits for the firm. In the other sort of task, the objective is to avoid
error or failure, because failure is tremendously costly to the organization.
Such tasks are called guardian tasks: Think, for instance, of the captain of a
tanker carrying a load of crude oil: Cutting a few hours off the time it takes
her to deliver her cargo is marginally beneficial to the firm, but incurring a
catastrophic spill is hugely costly to the firm; risking the second outcome to
achieve the first is, from the perspective of the organization, a very bad bet.
The question is, How do you set goals for an employee whose job mixes
both types of tasks? The satisfaction that comes from a brilliant success, if
you set that as a goal, is tremendously motivating. Indeed, the psychological
satisfaction that comes from brilliant success is probably such that no
explicit goal must be set in that direction. In any case, the achievability part
of SMART may be difficult to satisfy in such cases. And while avoiding
failure is probably achievable, doing so probably packs a much smaller
psychological wallop. For jobs that mix star and guardian tasks, goal setting
will be difficult, to say the least.
Goal-setting can be a good and even powerful motivational tool if you
can meet the requirements of SMART, plus challenge. But, especially for
Type-K jobs, this can be hard to do.
Equity Theory
Equity theory starts with the idea that people, and employees in particular,
have a natural desire to see that matters are arranged equitably or fairly. In
the context of employees, the theory supposes that an employee looks at the
contributions she makes to the organization and the rewards she receives, and
compares the ratio of these two for herself to the same ratio for others in the
organization, especially peers with whom social comparisons are natural. If
the individual perceives that her ratio of rewards to contributions is
markedly lower than the ratio for others, she is demotivated. She is also
demotivated (or engages in some reframing—keep reading) if her ratio is
markedly higher than the ratios of others. Indeed, the theory says that an
employee can be demotivated whenever she sees inequitable ratios of
rewards to contributions, even if it involves employees other than herself.
A textbook example of equity theory at work (certain to be featured in
future textbooks) concerns Gravity Payments of Seattle. Founder, CEO, and
majority shareholder Dan Price announced in April 2015 that he would raise
the minimum salary earned by any employee of the company to $70,000
(taking three years to reach that level). As reported in the New York Times,4
while reaction to the announcement was generally positive, a handful of the
most productive employees were less than enthusiastic: The Times article
reports that “Two of Mr. Price’s most valued employees quit, spurred in part
by their view that it was unfair to double the pay of some new hires while the
longest-serving staff members got small or no raises.” Employee Maisey
McMaster is quoted, “He [Price] gave raises to people who have the least
skills and are the least equipped to do the job, and the ones who were taking
on the most didn’t get much of a bump.” She made her views known, was
accused of being selfish and, already suffering from burnout, left the
company. Dissatisfaction wasn’t limited to the high earners: The article
quotes another employee, on the lower end of the salary spectrum, as saying,
“Now the people who were just clocking in and out were making the same as
me . . . It shackles high performers to less motivated team members.” And, a
final quote: “Am I doing my job well enough to deserve this? . . . I didn’t
earn it.”‡
In equity theory, it is the perceptions of individuals about the ratios of
rewards to contributions that matter. And, individuals have a lot of latitude in
how they define and measure both rewards and contributions. This latitude
can sometimes temper perceptions of unfairness. For instance, academic
accountants are paid quite handsomely compared to their peers in other fields
such as economics and marketing. But they do more or less the same work:
They teach classes and do research. This certainly rankles economics and
marketing faculty members, to some extent. But, at the same time, the belief
takes hold—among both accounting faculty members and their peers in other
fields—that teaching accounting is a particularly valuable and at the same
time onerous duty. This raises the perception of the contributions provided by
faculty members in accounting, and so brings the ratio of their rewards to
their contributions somewhat into line.
And, when employees perceive injustice, instead of being demotivated
they can be mal-motivated, taking actions like pilfering or padding expense
accounts: Someone who perceives that she is unjustly under-compensated
can respond by decreasing her contributions (which, essentially, is
demotivation) or by (illegitimately) enhancing her rewards.
Because it is individual perceptions that matter, and because those
perceptions are subjective, employers find, more often than not, that
perceptions exacerbate rather than temper feelings of unfairness. Suppose
that, in a particular organization, employees who are otherwise similar in
terms of education, tenure, and other socially-relevant characteristics
specialize in different tasks. To give an example, suppose Bob from Zephyr
Corporation is primarily responsible for supervising teams of locals who are
doing the actual construction, while Carl, who joined the firm at the same
time as Bob, has been assigned by Alice to work at keeping the clients happy.
It would not be unusual for Bob to perceive that, since he is involved in
actual construction activities, his contributions are more important than those
of Carl. Meanwhile, Carl is likely to perceive that anyone can direct work
crews, but it takes deft political skills to keep clients happy; his contributions
are more important and valuable. Now suppose that Alice has to determine
annual bonuses to give to Bob and Carl based on their “contributions.”
Whatever she does, either Bob or Carl will perceive their ratio as below that
of the other; if she gives them equal bonuses, they will both feel that way.
Another manifestation of this general phenomenon concerns the
distinction between effort and accomplishments. Ann may be endowed with a
level of skills greater than that of Bill, so that Ann can achieve a higher level
of contribution (measured, say, by the quality and quantity of product
produced) with a lower level of effort (measured, say, in terms of time spent
working, for salaried employees). If you pay Ann and Bill the same, Ann will
feel that her contributions are greater than Bill’s, so she is under-rewarded.
And, at the same time, Bill will feel that he is exerting himself more, so he is
under-rewarded. Good luck with that one! Going back to the case of Gravity
Payments, I bet that the reason the announcement met with generally positive
reviews from employees (and perhaps did more good than harm) is that
employees at the “base” of the compensation pyramid felt that they were
working just as hard as their better-compensated fellow employees; perhaps
the program provided the motivation to increase their contributions to the
organization. The feelings of injustice quoted in newspaper accounts came
from those employees who focused on (their perception of)
accomplishments, not effort.
And good luck dealing with Ann and Bill if they are salespersons, and
Ann sells more in a given period than does Bill, even though they work equal
hours. Ann will perceive this as a matter of her skill; Bill is likely to think
that Ann is “lucky” in the client base to which she was assigned.
Finally, and at the risk of going into too much detail on this one topic,
consider the commonplace system of merit-based raises in base salary.
Individuals who performed exceptionally well over the course of a given
year get, say, a 10% raise in their base salary, good performers get a 6%
raise, average performers get a 3% raise, and poor performers (who aren’t
fired) get only a 1% (nominal) raise. The problem with this is that it fixes
base salaries for the next round of raises. Now imagine Ann and Bill are
hired for the same job at the same time. Imagine they perform at the same
level every year except for year 2 of their employment, when Ann does better
than Bill. Then, while Ann and Bill get the same percentage raise in all years
except year 2, Ann is paid more than Bill in every year after year 2 (and, in
fact, gets a slightly larger dollar raise). Bill, looking at the ratio of his
contributions to his salary, and comparing with the ratio for Ann, finds this
entirely inequitable—Bill does just as much and as well for the organization,
but Ann, year after year, is paid more.§
This all sounds terrible: Employees are demotivated by perceptions of
injustice, and since perceptions are what matter, you, as employer, will have
a hard time getting it “right.” What do you do?
Psychology offers three suggestions. The first is based on the distinction
between distributional fairness (or justice) and procedural fairness. What
we’ve discussed so far is, in essence, distributional fairness: Never mind
how it happened, the employee looks at her ratio of rewards to contributions
and compares this to the ratio of other employees. Procedural fairness
concerns the process by which rewards are determined: Is this process
viewed (by employees) as basically fair and legitimate, even if it sometimes
results in distributional injustice? If you can devise a process for determining
rewards that is perceived to be procedurally fair and legitimate, and if you
can focus the attention of employees on the process and away from the bare
outcomes, this can temper feelings of distributional inequity. So, for instance,
if your organization is one in which turnover is low and long-term
employment is generally thought to be important to the organization, basing
rewards on tenure—insofar as it is viewed as legitimate—can work well.
(And, if you go down this route, you want to reinforce the legitimacy of
tenure as a measure of contribution in other, perhaps symbolic ways: For
instance, publicly recognize tenure anniversaries, with speeches about how
important are the “pioneers” to the success of the business.)
One important aspect of perceptions of procedural fairness (or the lack of
same) is transparency about the process by which rewards are determined.
Imagine an organization that hands out, say, year-end bonuses, admonishing
the recipients not to share what they are getting with their peers. Such
admonitions may not be effective. And, even when they are effective,
employees are likely to ask themselves, “What are they trying to hide?” This
doesn’t mean that you should necessarily publicize what each employee is
receiving; for one thing, doing this focuses employee attention on the
distribution of rewards. But if you have a process that (you believe, and you
believe your employees will believe) is legitimate and fair, publicizing the
process both is good in itself (“they aren’t afraid to tell us how this is done”)
and focuses attention on the process and away from the specific distribution.
Second, you can try to affect the comparison set used by your employees
to gauge equity. If you are rewarding different classes of employees
differently, steps—real or symbolic—taken to distinguish members of one
class from members of another can sometimes be effective. (When we get
back to Artisans’ Alliance, I provide some concrete examples of how this
might be done.)
And, third, remember that rewards include financial compensation, but
they can extend into other “goodies” that the organization can bestow. Keep
the survey data from last chapter in mind in this context.
Self-Determination Theory
In the theories explored so far, motivation is tied to external stimulus of
some sort or other: In expectancy theory, the employee consciously chooses
behavior that (she perceives) is desired by the organization, so that she will
receive from the organization a reward that she values. In goal-setting theory,
while individuals can set their own goals, the sort of employment-context
application I have in mind is where management sets goals for employees. In
equity theory, the individual employee is reacting to rewards provided by
and contributions provided for the organization.
Self-determination theory, in comparison, concerns motivation that is
intrinsic to the individual. The individual acts in a particular way because
she wants to do so, even absent external rewards or stimuli. The theory holds
that an employee is more likely to be self-motivated the greater her
Self-Perception Theory
Self-perception theory was described last chapter. To reiterate, it is based on
a process of retrospective justification leading to future behavior. The basic
notion is that individuals often act without having clearly defined objectives;
after the action is taken, the individual looks for a “story” that explains why
she acted as she did, and the story she adopts affects her future behavior. If,
for example, she justifies her efforts with the story that the work she did
serves some social goal about which she cares, then her future behavior will
reflect enhanced care about that social goal.
Recall the young engineers, designing the hardware and code for a new
minicomputer in The Soul of a New Machine. Perhaps, at first, they worked
long and onerous hours because of their enthusiasm for and excitement about
a brand-new machine. But one might expect their initial enthusiasm and
excitement to wane, as they continued to work under exhausting and stressful
conditions. How did they rationalize to themselves why they were doing
this? What kept them going? It wasn’t some promised monetary reward; they
weren’t going to be paid a big bonus if they succeeded. At least as Kidder
tells the story, they didn’t have a lot of affection for the firm for which they
worked. (They did, however, have affection for each other and the leadership
of their group, and peer pressure could well have been involved.) They
weren’t engaged in any sort of social mission. All that is left to explain their
efforts is that the work itself continued to be fun, interesting, and exciting.
And, if they perceived this was what had motivated them in the past, it
becomes a piece of their “identity”; they perceive that the work is fun, hence
they desire to continue to have the opportunity to do it. Indeed, this process
“manufactures” in them the desire to be able to continue to do this sort of
work once the current computer is designed; if the reward for successfully
working (under onerous and stressful conditions) on interesting and exciting
work is the continued ability to do so—Kidder’s so-called pinball effects—
this fills in the missing piece (the valence piece) of the expectancy-theory
story of their behavior.
Employees in some cases will have a choice of how they rationalize their
past behavior and, depending on their choice, we get different values, and
hence different behavior. Salience of a particular “story” makes it more
likely to be the chosen rationale; to the extent that management wishes to
employ self-perception theory to its own ends, it should determine which
story is the one it wants employees to embrace, and it should set conditions
to make that story the most salient.
For instance, in The Soul of a New Machine, Kidder describes how the
air-conditioning didn’t always work in the basement in which the young
engineers slave away, making their working conditions even more miserable.
And the author, based on conversations with more senior members of the
team, speculates that perhaps the boss of the team was sabotaging the air-
conditioning system; the more miserable the working conditions, the more the
team will band together in an us-against-the-world mentality, and the more
they will decide that the only reason they are working under these conditions
is because the work itself is interesting, exciting, and challenging.
Or, insofar as an organization wants employees to come to internalize the
welfare of the organization—employees are willing to sacrifice their self-
interests, at least to some extent, to help the organization succeed—the
organization wants employees to embrace a story of “I did it for the
organization, which I (therefore must, if this story is to make sense) value.”
An organizational culture that resembles a family, breaking down social
distinctions when they exist, emphasizing benefits with special consideration
given to those in particular need, and so forth, can make this specific process
of self-perception more salient.
Or, where the organization wants employees to believe that they are
willing to sacrifice themselves for some greater social good, making
achievement of that social good more salient—think of a hospital that, in its
communications with employees, emphasizes lives saved and patients cured
and de-emphasizes costs saved—will help in making this the story the
individual employee embraces.
What might top management at AA do? If they had a time machine at their
disposal, I’d recommend going back in time and looking for sales reps who
bought into the mission, signified by a willingness to accept nonstandard and
even subpar compensation packages. Absent a time machine—in other
words, for real—I’d recommend trying very hard to break the chain of social
comparisons. If it isn’t too late (and it may well be), contract with a third-
party firm to represent AA with giant retailers, a firm that in turn hires the
sales reps. Or, perhaps, find ways and means to compensate the Explorers
that indicate that they are different from the sales reps: For instance, offer
them (the Explorers) shares of stock or stock options in AA, with a clear
statement that this is for the “real” employees of the company—who will
benefit financially in the same way as will top management—and not for the
mercenary sales reps.
Expectancy theory: Motivation increases when expectancy (likelihood that certain actions lead
to what management wants), instrumentality (likelihood that fulfilling management’s desires will
lead to rewards), and valence (likelihood that those rewards are valuable to the employee) are
strong. Managerial implications: Set clear goals, with clear paths to achievement, and make
sure the rewards for achieving those goals are salient and valuable to your employees.
Self-determination theory: Employees are motivated by autonomy, the ability to learn and
exhibit skills, a sense of social relatedness, and a sense that work has a socially valuable
purpose. Managerial implications: Enhance these characteristics in tasks and jobs, and remove
barriers to feeling this way. (Of course, this takes us back to Chapter 1 and Robert Bass’s
principles of management.)
Self-perception theory: Employees, after the fact, look for reasons they did what they did.
And the explanations they arrive at are strengthened motivations for future behavior.
Managerial implications: To the extent you can, “steer” their reasons in directions that will align
their interests with yours.
Attribution, social comparisons, and motivation: Cathy attributes motives to Don’s behavior.
If Don is socially similar to Evelyn, Cathy may attribute similar motives to Evelyn. And if Cathy
is socially similar to Don and Evelyn, she may attribute similar motives to herself, guiding her
future behavior. Managerial implication: Build an organizational culture where employees are
similarly motivated, and a virtuous cycle will be created.
Undermining: For a variety of reasons that relate back to previous theories, providing extrinsic
rewards can dull intrinsic motivation. Managerial implications: Particularly for highly intrinsically
motivated employees, be careful (and perhaps simply avoid) piling on extrinsic rewards.
_____________
*
Beth Israel merged with Deaconness Hospital in 1996, forming BIDMC, the Beth Israel Deaconness
Medical Center. Notwithstanding this, to keep the narration simple, I will use the names “Beth Israel”
and “BI” for this organization both before and after the merger.
†
To keep the pronoun references clear in this story about Beth Israel, I use the somewhat sexist she
and her for nurses, he, his and him for doctors, and he or she for patients. In addition, generic
employees for this this chapter will be she.
‡
Since I wrote this account of Gravity Payments, allegations have been made about why Price decided
to set the $70,000 minimum wage. See, for instance, Todd Bishop, “Gravity Payments Defends Business
Practices Amid New Allegations of Financial Trickery,” Geekwire, February 1, 2016. This is an ongoing
story, and to see how it all ends, I can only suggest that the reader search the Web for the latest
developments.
§
Stanford GSB employs merit-based raises and, from my decade as senior associate dean (with the job
of listening to faculty colleagues complain about their salaries), I can assure you that this is no make-
believe story. For more on the disadvantages of merit-raise systems, see Baron and Kreps, Strategic
Human Resources: Frameworks for General Managers (New York: Wiley, 1999), pp. 258–61.
¶
In part, the controversy exists because Titmuss offered little empirical evidence in support of his
contentions. A relatively recent study gives some support to Titmuss’s predictions; interestingly, they find
that compensated donations diminished primarily among women, and there was no decrease in donations
when subjects in their study are allowed to donate their payments to charity. C. Mellstrom and M.
Johannesson, “Crowding Out in Blood Donation: Was Titmuss Right?,” Journal of the European
Economic Association, Vol. 4, 2008, 845–63.
7
The discussion so far has largely focused attention on the motivation of the
individual employee. How does all (or any) of what we’ve said apply, when
individual employees work within teams?
1. One size does not fit all. Team motivation depends entirely on
what the team is being asked to do, as a team. “One size fits all” is
almost never true when it comes to motivation; team motivation is
a domain in which “one size fits all” doesn’t even make sense.
2. Beware the free rider. When the issue is that measures of
individual performance are lacking, so any contingent rewards
must be based on joint-performance measures, the free-rider
problem arises. Peer pressure within the team can be a manager’s
friend in such cases, but watch out for peer pressure that works
against management’s interests.
3. Rely on psychological rewards. Psychological rewards rather
than pay-for-(team)-performance rewards are generally
advantageous in these settings, because they need not be “divided”
among team members.
4. Boost reciprocal cooperation. When the issue is having team
members assist one another with their privately assigned tasks,
look to engage reciprocal cooperation that emerges “organically,”
rather than trying to “make cooperation happen” by imposing a
reward scheme.
5. Manage the brainstorm. And when it comes to teams that are
tasked with collectively arriving at a good solution, management
of the team process is important; time to add to your list of things
to read.
_____________
*
Also, understand that, even in this case where no environmental uncertainty enters—where the
completion time depends only on the effort level of each of the three—the incentive scheme “get it done
on time or you are all fired” can have a dark side. Let’s say Dorothy’s level of effort can substitute for
Bob’s and Carl’s—that is, if she works harder, they can slack off and still make the deadline set by
Alice. If Bob and Carl are jerks, they can tell Dorothy, “We aren’t going to work very hard, so it is up to
you to make extraordinary efforts to keep us on schedule. And if you don’t, you’ll be fired along with
the both of us.”
For those who know the lingo, everyone working as hard as Alice wants is one Nash equilibrium, but
it is only one of many Nash equilibria in this situation; this situation is essentially equivalent to a classic
bargaining game, where parties have to decide what “demands” to make; if the sum of their demands
are less than some fixed amount, each gets what he or she demanded; if their demands sum to more
than this amount, everyone gets zero. And, as both theory and practical experience tell us, in such
situations, a lot of outcomes—many involving the majority ganging up on a minority—are possible.
†
A classic paper in the economic theory of incentives (Bengt Holmstrom, “Moral Hazard in Teams,”
The Bell Journal of Economics 13 (1982) 324–340) develops this point.
‡
Two cheap, but not necessarily trivial, “process” suggestions about this are: (1) Don’t ask, “Who came
up with that idea?” Instead ask, “How did the group arrive at that answer?” and (2) Encourage the
soccer practice of honor-sharing. Encourage a culture in which Ann, if she is identified as the author of
the idea, points out whose contributions inspired her, and so forth.
§
When I teach a case study in courses, I often take advantage of this phenomenon by asking students
to vote, before our discussion, on which course of action they think is right for that particular case. In a
case discussion, I’m trying to get the students to explore all the options, and getting them “committed”
up front to a particular answer ensures that, even if one option is a loser, its death in conversation will
take place only after all its flaws (and virtues) have been hashed out.
8
Perhaps you picked up this book in the hope that it would provide an easy,
specific plan to improve the motivation of employees at your organization.
As I said at the start, I can’t provide any concrete and definite
recommendations how you might do this, because such recommendations
necessarily depend on a host of specifics about your organization that I don’t
know. Instead, what I’ve tried to do is to give you some tools for thinking
about motivation, in the hope that those tools, combined with your
understanding of your own situation, will lead you to better motivation.
I’ve used the word “tools” whenever possible to describe what I’m
offering; perhaps now, after you’ve made it nearly to the end, I should
acknowledge what is probably already obvious to you: The tools provided
are of two distinct types. First, I’ve described and discussed specific
motivational “tools”—“schemes” might be a better term—such as pay for
performance and goal-setting. But I’ve also offered conceptual tools or
frameworks, ways of thinking about your relationship with your employees
and their relationship with you, with your organization, and with their work.
If applied with insight, these “thinking tools” will help you to understand the
advantages and disadvantages of specific motivational strategies in your
specific context. Kurt Lewin, one of the pioneers of social psychology and its
application to management, famously said “Nothing is as practical as a good
theory.” As a management educator, I subscribe to this, although I would add
to his aphorism: “Nothing is as practical as a good theory intelligently
applied.” I hope that the theories and frameworks I’ve provided are good.
But it is up to you to apply them intelligently to your specific context.
This doesn’t mean that there isn’t generally applicable advice that I might
give. I believe that organizations, and the people who run them, tend to have
certain categories of “blinders” on when it comes to motivation. I’ve
mentioned some of those in passing in previous chapters and, at the end of
this chapter, I’ll summarize the four items of significant general advice that, I
believe, apply to many and perhaps most organizations.
But, before doing that, I would like to offer a different sort of assistance.
Suppose you hired me as a consultant, with the general task of suggesting
how to improve the motivation of your employees. The first thing I would do
is ask you to answer a series of questions about your organization and its
overall situation and environment. So, in this chapter, I’ll go down a list of
questions I would ask, together with explanations for why I think answers to
those questions are key and how I might use some answers to provide you
with advice. This is not a complete list of the questions I would ask you.
Often, the answers you might give would inspire follow-on questions. But if
you have absorbed the frameworks and ideas of this book, I believe that
these questions, honestly answered by you after serious contemplation, will
point you in the right direction.1
You might think answers to this question are so obvious that no retrospection
is required. But, in my experience, being thoughtful and explicit about the
link between strategy and employee efforts provides a lot of insight. Recall
the story from Chapter 2 about the chain of health-food lunch shops, where
managers at two different stores, given the incentive to raise the bottom-line
results of their store, went in completely different directions. In the
discussion back in Chapter 2, we took it as given that this was a bad outcome
for the organization as a whole. But was it? This depends on the strategy of
the organization as a whole: Is the chain’s overall brand image important to
it? Or would it be perfectly fine for the store in one location to have an
entirely different look and feel from another location? Consistent brand
image across locations is certainly important when a large proportion of the
clientele at any one location are “drop-ins,” brought into the restaurant
location by some consistent brand image. But suppose instead that most
customers in a location are customers only in that location and, moreover, are
“regulars,” who come to know the restaurant in their location from their
experience in that location. Then the chain might well prefer that employees
(in particular, managers at each location) undertake efforts to tailor what they
do to the locals; if, say, one location serves primarily business people from a
local office park, while another serves students from a nearby university, it
may make sense to encourage distinct identities.*
Begin by identifying your organization’s strategy. You can use whatever
framework works for you, but if you want the framework that is taught to
MBA students at Stanford GSB,2 it involves answers to the following four
questions:
4. What is the logic that ties all this together? Why will the organization
succeed in achieving its goals?
And, since we’re concerned with employee motivation, add to these a fifth
and sixth question:
6. How do your HRM policies and practices elicit from your employees
the desired behaviors and actions? How well do they do this?
The economic environment in this question is not the market in which your
products or services compete—those things enter into your business strategy
—but instead the marketplace for labor services that you face and, then, the
economic environment that faces your (prospective) employees. Contrast, for
instance, the situation facing SAS Institute,3 an enterprise-software firm
located in Cary, North Carolina, with that facing one of its competitors
located in Sunnyvale, California, in Silicon Valley. It is hardly true that Cary
is located out in the “sticks”; it lies inside the Research Triangle formed by
Durham, Chapel Hill, and Raleigh. But job-hopping—lateral movements by
employees from one firm to another—is much less prevalent (and possible)
in North Carolina than in Silicon Valley. Hence, SAS can and does use
longer-term motivators with its employees. It invests heavily in its
employees—providing “general-skills” training—investments that pay a
return for SAS, with its very high retention rates, that is higher than would be
the case for a firm in Silicon Valley, which might even lose an employee after
providing general-skills training because that training makes the employee
more desirable for other local firms. And you should also consider the
condition of other markets in which your employees deal: The housing
markets in and around Cary are very different from those in and around
Sunnyvale; this can make quite a difference in how firms in the two locations
structure benefits for incumbent employees and moving allowances for
prospective employees from outside the region.
As for the social environment, consider for instance HRM practices that
celebrate and recognize a single “Employee of the Month,” meant both to
motivate workers and to provide them with a concrete model of what sort of
behavior is desired by the organization. In a culture that celebrates individual
achievement—say, in Boston—this would fit a lot better than it would in
Southeast Asia, where individuals strive not to stand out from their peers. (I
have heard an anecdote, perhaps apocryphal, that when implementing an
“Employee of the Month” program in rural Thailand, a company found that
the unlucky winner was shunned and abused by his or her peers. The program
was quickly discontinued.) Or, think of the differences in the social cultures
of North Carolina and Silicon Valley and how those differences manifest
themselves in the behavior of employees at SAS Institute versus, say,
employees at Google.†
I doubt that I need to emphasize that HRM policies and practices in
general need to fit with or, at least, take heed of the local market conditions
and social environment. But general managers rarely appreciate the
complexity of labor laws, at least in developed economies. Labor law in the
United States is vast; moreover, it varies by state, with municipalities and
cities sometimes adding more statutes related to worker rights. Of course, the
elephant in the room—although it is found in increasingly fewer rooms—is
organized labor; in the United States, organized labor (outside of certain
geographic and industrial settings, and outside of the public sector) is
unlikely to be a factor for you, but if your organization has a presence outside
the United States, it can be significant. But, beyond organized labor, laws and
regulations on what and how you pay employees, which benefits can be
offered on a tax-preferred basis, and—with the advent of the Affordable
Care Act and, as I write these lines, with whatever comes next—the rules
around health care are many and complex.
Here are two specific examples that touch more or less directly on
motivation:
Please note that I’ve phrased this question as, “What (if anything) is
special . . . ?” I imagine that, for a lot of employers, the first reaction to this
question is, “Nothing much is special.” But, don’t be fooled; instead dig
deeper. There is probably more there that impinges on you than you initially
realize.
Please note that, in this classification system, it is the job and not the
employee that is star, or guardian, or foot soldier. “She is a star employee” is
a perfectly good English sentence, presumably meaning that she does her job
in exemplary fashion. If she is driving a chemical carrier for the organization
and does so with a spotless driving record while having the best on-time
record of any driver in the company, she is a star truck driver. But her job, in
this classification system, is still a guardian job.
And this matters because in terms of good HRM practice, what you do to
recruit, train, monitor, evaluate, and motivate someone in a star job varies
from how you do these things for a guardian job. For instance, in recruiting
someone for a guardian job, you probably want to be very careful in terms of
the individual’s past work performance and references. You probably want to
train the individual extensively. You want to motivate the person not to fail:
The truck driver might be given some incentive to be on time, but it should be
mild incentive and should allow for explanations of why the driver was late.
For a star job, each of these is turned on its head: I wouldn’t go so far as to
say you want to seek out employees who will “gamble” in their efforts and
then provide them with the motivation to do so, risking the strong likelihood
of a bad outcome for a small chance of a very good one. But that isn’t far off.
And, in some (usually the most problematic) cases, job designs are mixed
star-guardian jobs, where exceptional performance on either end of the
scale has huge consequences for the organization. Service jobs that call for
creativity and proaction, but where the organization’s reputation with its
customer base is also important, often have this bad mix: Exceptional service
can earn a lot of profit for the firm, but the employee who spends a
disproportionate amount of time looking for the “home run” and so strikes out
(screws up) the service of even a few customers can be a disaster.
Employees are put in jobs that are mixes of tasks. What are those mixes, and
why have you created jobs with those mixes? As just said, jobs that mix star
and guardian tasks are problematic. Back in Chapter 2, we discussed why
motivation can be difficult for employees whose jobs mix tasks with quick-
and-easy-to-measure performance and tasks for which it is difficult to
measure performance, at least in a timely manner. There may be good reasons
to bundle together tasks that mix these qualities into a single employee’s job
because the tasks are complementary, perhaps because of the employee’s
possession of relevant information, perhaps due to production technology.
But you should not take job designs as given, and you should certainly be
clear on the full list of tasks given to an employee before you set out to
design his or her motivation. And if you do have job designs that mix hard-
to-assess-performance tasks with easy-to-assess-performance tasks, this
should certainly incline you toward the more psychology-based motivational
tools.
This set begins with classic demographic characteristics: What is the age
distribution of your employees? What are their educational backgrounds?
How many are married and with children? How diverse are they, along
multiple dimensions, including race, creed, gender, and ethnicity? Few
HRM policies and practices are universal in the sense that they work well
for every workforce, and you want your practices to be well suited to the
people who work for you. So, obviously, you want to know who they are.
Having said this, I must immediately issue a warning: Discrimination
based on these demographic categories, either in hiring or in employment
practices, can land you in legal hot water. When it comes to job interviews
or even information gathering about current employees, certain questions
cannot be asked. There are cases where demographic discrimination in hiring
is permitted; when membership in a demographic group can be shown (in
court) to be a bona fide occupational qualification. But those cases are hard
to make; an understanding of the law, which may require consultation with a
good labor lawyer, is essential.
And, beyond your legal obligations: While “you want your HRM
practices to be well suited to the people who work for you” is good advice,
it can be dangerous advice if you, like most people, are subject to prejudices
about the tastes and behaviors of different demographic groups. Such
prejudices can seem scientific, based on statistics properly applied to
samples of past workers who are relatively homogeneous.§ But often they are
based on the blind application of faulty folk wisdom.
Consider, for instance, Sun Hydraulics, a firm that manufactures and
designs fluid-power devices (hydraulic valves and manifolds),
headquartered in Sarasota, Florida.¶ Sun was founded by Robert Koski, who
held contrarian views about the behavior and motivation of skilled blue-
collar workers, among other management matters. The production technology
for producing hydraulic valves and manifolds is pure job-shop: Highly
skilled blue-collar workers use general-purpose machine tools to convert
raw materials into finished products, involving for each product a complex
sequence of steps that varies from product to product. Folk wisdom is that
job-shop operations live or die by (a) scheduling work skillfully, which is,
per folk wisdom, beyond the capabilities of the blue-collar fabricators, and
(b) closely supervising the blue-collar fabricators, again based on folk
wisdom about the behavior of this class of employees. Koski believed that
this was hokum, that his blue-collar workers, treated with the respect they
deserved both for their skills and for their knowledge, could do a much better
job if provided with all the information they needed to self-schedule and
self-supervise. (“Self” here may be a bit of a misnomer: Peer pressure
among the blue-collar workers at Sun is probably ferocious. The point is that
white-collar scheduling and supervision is at least unnecessary and perhaps
suboptimal.) The history of Sun certainly indicates that Koski was right and
folk wisdom about the nature of high-skill blue-collar workers was, at least
in this case, wrong. And lest you think this is something only true about
highly skilled blue-collar workers, ServiceMaster is a successful company
founded on HRM principles very similar to those of Sun, but for semi- and
unskilled blue-collar workers.#
Compiling demographic statistics about your workforce is important, but
it is just the beginning of answering the question, Who are your employees?
Go back to Chapter 5. Are your employees’ connections to their work and
their jobs instrumental or expressive? If they are expressive, what is the
nature of the expressive connection? Are your employees work driven?
Values driven? Mission driven? Driven by the success of the organization?
Cutting this along a somewhat different dimension, do your employees regard
their relationship with your organization as a job, a career, or a calling?
One way to understand what Koski achieved at Sun Hydraulics employs
these categories. The folk wisdom about the behavior of blue-collar workers
may be accurate to the extent that blue-collar workers have a largely
instrumental connection to their work and to the organization for which they
work. The blue-collar workers at Sun, however, have a much stronger and
expressive connection to the work they do and to the organization for which
they work, which changes how they behave on the job; Koski’s insight was to
understand that building this sort of connection was possible by treating this
group of employees with respect and by giving them autonomy over their
day-to-day activities.
I doubt that any set of questions, except perhaps for the question linking
business strategy to desired employee behavior, are more deserving of deep
reflection than questions about how employees connect psychologically to
what they do on the job. The response of your employees to different
motivational tools you can employ will be profoundly affected by the
answers. The sad story of Artisans’ Alliance is another good case in point;
for still another, consider furniture maker Herman Miller.5 Herman Miller
(HM) is best known for its office furniture, including such iconic pieces as
the Eames chair, the Aeron chair, and the Action Office modular office
system. Manufacturing takes place in Zeeland, Michigan, with a workforce
that is strongly evangelical Christian in the Reformed Protestant tradition.
The firm celebrates these values, honoring employees for and giving them
paid time off to engage in community service. Employees at HM are at least
values driven, and a case can be made that they are mission driven, where
the company’s mission statement is “Inspiring designs to help people do great
things.” All this has resulted in employees who are able—and willing, and
expected—to take part in the day-to-day management of factory-floor
operations.
For a variety of reasons, HM decided in 2012 to undertake a major
strategic initiative. This included expanding the scope of operations, both in
terms of product and, importantly, geography. Expansion was achieved
through acquisition, and geographic expansion involved acquiring a Chinese
office-furniture manufacturing company, POSH Office Systems. POSH is
described as a “very traditional Chinese company,” which means, with
employees whose relationship to their company is very different from the
relationship of HM and its core employees. As you would expect, the
acquisition of a company with such a different employee-employer
relationship raises issues of whether HM should try to keep POSH as it is or
remodel it along HM lines and, if it decides to change POSH, how and how
fast. Making these decisions, in turn, requires that HM understand how POSH
employees are attached to their jobs, their work, and their employer.
But it is not HM’s relationship with its new employees at POSH that
motivates the telling of this story. Instead, consider how HM’s actions at
POSH might affect its relationship with its core workforce back in Michigan.
Those workers share values and even a mission with HM. Suppose they
misinterpret what HM is doing in China. Suppose HM does not try to change
the employee-employer relationship at POSH, and the employees back in
Zeeland recognize this and decide that this is not in accord with the values
they thought they shared with HM. Or suppose that the employees in Zeeland
are simply kept guessing about HM’s intentions in China. In any of these
circumstances, the psychological contract between HM and its core
workforce—which is worth a great deal to HM in terms of management and
motivation—could be badly damaged.** This is an ongoing story, so we don’t
know how it will end. But it should be clear that top management at HM, in
deciding what to do in China, must consider the impact on employee attitudes
back home. And they must go out of their way to keep their core workforce
informed and “on the same page” concerning the strategic initiative and the
details of its implementation.
This story suggests the next set of questions you should ask yourself:
QUESTION SET VI. Are you and your employees on the same
page?
If your employees were asked the questions in Question Sets I through III,
would their answers be consistent with the answers you gave? In other
words, are you and they on the same page concerning: (1) your business
strategy; (2) their role in that strategy; (3) their relationship with you, your
organization, and their jobs; and (4) the details, explicit but especially
implicit, of how employment “works” at your organization? And, how do
they feel about these things? In particular, do they believe that your policies
and practices are equitable and, where they do not believe this, what is the
source of their dissatisfaction?
If your employment relationship is, for your employees, a one-off,
instrumental transaction—they work, they get paid—then you can probably
make do with relative ignorance on their part, as long as they understand
what they are meant to do and how they will be compensated. But if you have
the sort of long-term relationship described in Chapter 4, and especially if
you have a psychological “contract” with your employees in the sense of
Chapters 5 and 6, you are courting trouble if their understanding doesn’t
match yours. And, in many cases, employers are surprised by how poor is
their employees’ understanding of what the employer was sure is common
knowledge.
Obviously, you want to communicate these things to them. But you should
also listen to them, in direct conversation and via anonymous feedback
platforms.
One more question connected to Who are your employees? should be
addressed:
Are your HRM policies and practices attractive to the “right sort” of
prospective employees? Are they attractive to the “wrong sort?” (In both
cases, how and why?) To what extent are your HRM policies and practices
appreciated by prospective employees? To what extent do your recruiting
practices present an honest and complete picture of what employment with
your organization means?
These questions may be difficult to answer, because you don’t have ready
access to prospective employees who do not become employees. You can
interview your (newly arrived) employees and ask them, “What did you
know about us when you were thinking of coming to work here?” But, of
course, the fact that they chose to work for you makes them a biased sample.
You’d like to ask those who chose to work somewhere else, “Why?” And,
hardest of all, you’d like to ask those who chose not to apply at all, “Why
not?” But how do you do this? You could ask a consumer-marketing survey
firm (or your own consumer-marketing survey people, if you employ such
specialists) to run a survey that asks, “Have you ever considered working at
[insert name of your organization]? If not, why not? And what do you know
about working conditions there?” But will the data be worth the cost?
One indication that you need help on this front is the percentage of new
hires who quickly become ex-hires. Exit interviews with voluntary quits,
and, in particular, short-term voluntary quits, can be very enlightening.
Is there an overall organizational culture and, if so, what is it? If there are
separate cultures among different groups of employees, what are they? How
do your motivational schemes affect the culture(s), for good and for bad?
Organizational culture is a summary of how the members of your
organization relate to their work, their jobs, the organization, each other, and
to other constituencies of your organization, including suppliers and
customers. In a sense, this question, together with the question about your
business strategy, are bookends for all the other questions. The questions
about business strategy concern what you and your organization are trying to
achieve, and why you think you can achieve those things; those questions bias
you toward thinking of your organization as an entity with a purpose. On the
other end of the spectrum, you should think of your organization as a
“marketplace” in which individuals—the focus here is on your employees—
exchange economic and social goods. Your organizational culture aims at a
succinct description of the way in which those individuals relate to these
exchanges.
Often, an organization doesn’t have a single culture in these terms; but
instead different groups of employees have distinct cultures. For instance,
based on my personal experience, tenure-track faculty members and adjunct
faculty members at Stanford GSB have different cultures, and the many
different types of support personnel—clerical support, the development and
alumni support teams, the student-service teams—each have cultures with
some distinct elements. That’s not necessarily a bad thing; it is in fact a
natural outcome given differences in who they are in terms of demography,
education, and social background; what they do; the labor markets they face;
and so forth. But in thinking about motivating your employees, it is important
that you acknowledge these differences and, in particular, the role that your
motivational schemes play in creating, supporting, and, sometimes, grinding
against the culture or cultures within your organization.
After answering all the previous questions, you may find the answer to
this question to be, “There is no succinct description of the organizational
culture or the various subcultures within my organization.” In my view, this
answer should raise a red flag. At least when it comes to Type-K employees,
I would read the “no easy answer” response as a signal that, in economic
terms, your employment relationships are not constructed on a solid base of
mutual understanding and, in social-psychological terms, your employees
probably have tenuous connections to their jobs and the organization. While
I’m not saying that “no easy answer” necessarily implies these dangers, it
worries me, and I think it should worry you.
Intangible Contributes
Tangible Interesting Contributes to
personal to
rewards; & exciting greater social
rewards; organizational
e.g., pay work purpose
e.g., praise success
Intangible Contributes
Tangible Interesting Contributes to
personal to
rewards; & exciting greater social
rewards; organizational
e.g., pay work purpose
e.g., praise success
Not at all
0.5% 0.0% 0.0% 0.0% 1.0%
effective
Limited
6.3% 3.9% 0.5% 1.0% 10.6%
effectiveness
Extremely
27.5% 32.4% 58.0% 52.7% 24.6%
effective
Only this is
1.0% 4.3% 7.2% 7.2% 3.4%
effective
Standard
0.95 0.71 0.76 1.07 0.90
deviation
Not at all
0.0% 0.0% 0.0% 0.0% 0.5%
effective
Limited
8.7% 1.0% 1.9% 2.9% 20.3%
effectiveness
Extremely
21.3% 27.5% 51.2% 30.9% 13.0%
effective
Only this is
1.0% 1.0% 4.3% 1.4% 0.5%
effective
Standard
0.91 0.75 0.74 0.83 0.97
deviation
Table 4. MBA-Student Responses to the Five-Motivational-Channel
Questions
Intangible Contributes
Tangible Interesting Contributes to
personal to
rewards; & exciting greater social
rewards; organizational
e.g., pay work purpose
e.g., praise success
Intangible Contributes
Tangible Interesting Contributes to
personal to
rewards; & exciting greater social
rewards; organizational
e.g., pay work purpose
e.g., praise success
Not at all
0.4% 0.0% 0.4% 2.1% 5.0%
effective
Limited
7.5% 2.9% 0.8% 15.4% 13.8%
effectiveness
Extremely
26.3% 52.5% 65.8% 20.8% 22.9%
effective
Only this is
1.3% 2.9% 8.3% 1.3% 3.3%
effective
Standard
0.94 0.81 0.76 1.08 1.21
deviation
Not at all
0.4% 0.0% 0.0% 2.5% 5.0%
effective
Limited
4.6% 1.7% 1.3% 19.6% 26.3%
effectiveness
Extremely
33.8% 40.4% 49.2% 14.6% 10.4%
effective
Only this is
1.7% 0.4% 4.2% 0.0% 0.4%
effective
Standard
0.95 0.74 0.83 1.02 1.10
deviation
And not only that: Look at the percentages of respondents who say that
“interesting and exciting work” is at least “very effective”: for the SEP
participants about themselves, 94.2%; for the SEP participants about their
direct reports, 93.2%; for the MBA students about themselves, 92.4%; and
for the MBA students about their peers, 89.2%. This shouldn’t be surprising:
Giving someone work that they find interesting and exciting is, by itself,
powerfully motivating; it is the motivational magic bullet.
There are three obvious problems with this finding: First, sometimes
work must be done that is neither interesting nor exciting. Second, employees
who find a particular project interesting and exciting may take things too far,
spending too much of their time on that project when there are other tasks to
do.‡‡ Third, and another way in which things can be taken too far, employees
may keep working on an interesting and exciting project past the point where
a good-enough job has been done. These problems all come down to your
employee having too much motivation to do what he or she finds interesting
and exciting, spending more time and effort on that task while avoiding or
shirking other tasks.
These are real problems. But in my experience, with a Type-K employee,
a good motivational practice is to sit down and try to identify tasks that (1)
he or she would like to do, and (2) that will benefit the organization if done
well. When you can find tasks that satisfy both these criteria—and they exist
more often than you might think, if you and your employee look hard for them
—the motivation to provide consummate effort will flow nearly effortlessly.
Motivation in this book has been formulated as “aligning the interests of the
employee with his or her employer.” “Aligning” in this formulation suggests
that it is something you must actively do; that in the natural state of affairs,
your employees’ interests will not be aligned with your own, and you need to
do something about that.
It isn’t necessarily so. And if it is so, it may not be necessary for you to
do much. Especially when it comes to Type-K employees, their intrinsic
motivation is often to do good work and be seen doing good work, to
contribute to the success of the team, and to be respected and even admired
for what they do. These intrinsic motivations may be enough, and if you pile
on motivation by other means, you can dull the naturally occurring intrinsic
motivation.
Of course, your employees need to know what they should be doing to
help the organization achieve its goals. It helps if you take steps so that the
individual employee internalizes the success of the organization as his or her
own success; to the extent possible, you should promote the notion of the
employee as teammate or “owner.”
But always consider that in the context of motivation, less is sometimes
more.
If rewards are needed, think broadly in terms of what rewards you can
provide. Go back to Table 5 and the notion that intrinsic rewards may work
better on your employees than you (at first) think. Consider goal-setting and
achievement of a SMART goal as a reward. Consider self-determination
theory and psychological motivators in general. Sometimes, money—or the
prospect of promotion—is the most effective tool. This is particularly true in
organizational and social cultures where the rule is that “he who dies with
the most toys, wins.” But that is not a universal rule; it is probably less of a
rule than most managers believe.
One suggestion that I’ve made several times in this book is to always think
about the longer-term and broader-scope implications of how you are
motivating your employees. Too often, the demands of the here and now or
this specific case overwhelm longer-term and broader-scope considerations,
to the detriment of the organization. Assuming you have a long-term
relationship with the employee with whom you are dealing today, and
assuming that what you do with or about or for this employee will have
repercussions for your relationship with your other employees, you need to
be as mindful of the future as you are of the present.
And finally:
I return to Chapter 1 and the wisdom of Robert Bass. Management is, in most
contexts, first and foremost about getting things done through the efforts of
others. Rather than trying to control your employees through specific dos and
don’ts, your job is to find people who can excel and unleash their passion
and creativity in productive directions. Align their interests with yours—
motivate them—and you will reap the rewards together.
_____________
*
Although it is a bit off the point, I should add a follow-up question for this specific case: If the business
strategy doesn’t require consistency across different locations, what in the business strategy links
different locations? To give a “for instance,” it could be centralized procurement of raw materials.
Whatever the answer, it is important to get it out on the table, because it may still have implications for
what behavior is desired from employees.
†
But—always a but—it should be noted that as Google started out, top management visited SAS
Institute to see what they could learn from how SAS conducted its HRM. Google didn’t copy SAS’s
policies and practices precisely, but adapted the good things that they saw at SAS to the different social
and economic environment of Silicon Valley.
‡
The doctrine of disparate impact says that if actions you take affect different groups of employees
differently, based simply on statistics, and, in particular, if “protected classes” of employees are
adversely affected, those actions are discriminatory, even if you had no discriminatory intent. So, for
instance, if you provide one group of employees with greater incentive compensation than a second
group, because motivating members of the first group is more important to your bottom line, if the first
group’s average compensation is higher than the average for the second group in consequence, and the
first group is disproportionately populated with men and the second group with women, you may be
guilty of discrimination based on this doctrine. Or if you provide greater incentive payments for tasks
that require greater upper-body strength, which will (probably) create adverse outcomes for women,
you may be guilty based on this doctrine. As I write this, your defense in court is to show “business
necessity” of the practices that led to the disparate impact; in both stories I’ve told, if your incentive
schemes are economically sound, you might be okay. However, legal standards of what constitutes
“business necessity” change with time and, I think it fair to say, have tended to become harder to meet.
§
The problem here is one of sample size. If the data you process about, say, voluntary quits by men and
women is based on historical records where you’ve employed many more men than women, and you
are looking to fill a job where you want to be relatively sure that the person will not leave, the small
sample size of women will mean less assurance in your estimate. For more on this, see Baron and
Kreps, Strategic Human Resources: Frameworks for General Managers (New York: Wiley, 1999),
p. 353ff.
¶
Sun Hydraulics, a fascinating company, is the subject of a superb series of Harvard Business School
cases. If I can convince you to study the details of one company, to expand your ideas about employee
motivation (and employee behavior, more generally), it is to study Sun. To get started, consult Louis B.
Barnes and Colleen Kaftan, “Sun Hydraulics Corporation (A and B) (Abridged),” Harvard Business
School Case 9-491-119. But don’t stop there—in particular, see follow-up cases about how Sun dealt
with the recessions of 2001 and 2008.
#
Look on the Web for the page that describes ServiceMaster’s organizational culture,
http://www.servicemaster.com/about-us/we-serve.
**
Sales per employee in 2011 at HM were around $250,000, while sales per employee at POSH were
around $40,000. Of course, this doesn’t tell the whole story: Presumably, the levels of capital intensity
are very different, these are measures of average and not marginal productivity, and employee salaries
are bound to be different. But the scale of the difference is striking, indicating that HM should be very
concerned with its relationship with its core workforce.
††
In biology, the adjective punctuated refers to the notion that, in a formerly stable ecology, changing
one aspect can disrupt the ecological equilibrium, causing a rapid burst of evolutionary activity until a
new equilibrium is reached. Hence, evolutionary activity is marked by long stretches of relative stability,
punctuated with rapid bursts to accommodate any changes. In this setting, where the “equilibrium” can
be changed by an act of management, the term punctured equilibrium might be more suitable.
‡‡
Jim Baron, Diane Burton, and Michael Hannan conducted an empirical study of high-tech start-ups in
Silicon Valley, in which (among other things) they looked at how the “HRM vision” of the organization
correlated with firm performance. The dimensions of HRM vision that they used were only indirectly
about motivation of the workforce but, reading a bit between the lines, what they called a “star
organization” could reasonably be identified as an organization in which “do great technical work” was
the chief motivator. And they found that, while these organizations did not do best in terms of economic
and financial performance (versus other types of organizations), they did do best if one looked only at
those start-ups that made it to the stage of initial public offering. I believe that we are looking here at a
variant of the well-known adage: “the perfect is the enemy of good.” When, in these organizations,
motivation chiefly stemmed from doing a great (technical) job—when motivation was via interesting and
exciting work—the engineers and scientists who were “inventing” the product would often ignore
market realities to get to the most beautiful and scientifically excellent product. By looking at firms that
made it to IPO, they filtered out of their sample the beautiful failures, at least to some extent. Moral:
Interesting and exciting work is a great motivator. But not when it leads employees to ignore the
economic reasons for the work being done.
A good, nontechnical summary of this research can be found in Baron and Hannan, “Organizational
Blueprints for Success in High-Tech Start-Ups: Lessons from the Stanford Project on Emerging
Companies” (California Management Review, Vol. 44, 2002, 8–36). If you have any interest in HRM
in start-ups and, in particular, in high-tech start-ups, you should read this wonderful article.
APPENDIX.
THE WISDOM OF CROWDS: WHAT DO
MANAGERS BELIEVE?
8. For the people who report directly to you, which of the following statements is MOST
descriptive of what motivates their best work?
My direct reports do their best work when they can see a clear connection with tangible
rewards for themselves, such as a bigger bonus, higher pay, etc.
My direct reports do their best work when they can see a clear connection with intangible
rewards, such as praise and/or enhanced status and respect among co-workers.
My direct reports do their best work when they find the work interesting and exciting to
them personally.
My direct reports do their best work when they understand that the work is important to the
success of the organization (or work group).
My direct reports do their best work because they consider their work to be an important
mission that transcends their own personal interests and rewards and the well-being of their
associates and the organization for which they work.
9. Of course, people differ in what motivates them, and your answers to Questions 7 and 8 may
not do justice to the variety of motivations of your direct reports. In this respect, which of the
following statements best describes your situation?
My direct reports tend to be quite different in what motivates their best work. I find that I
must vary the techniques used to motivate my direct reports to get the best results.
While my direct reports differ in what motivates their best work, I find that I can (and do)
tend to motivate them in largely similar ways.
My direct reports are very similar in what motivates their best work, so I can treat them the
same when it comes to motivating them.
And then came Part III, concerning the eight types of motivator discussed
in Chapter 5.1 This began with a preamble:
The final part of the survey concerns eight general categories of motivator, slicing things
differently from in Part II. The eight categories we consider in this part are:
1. Benefits (such as health care, retirement benefits, etc.)
2. Feeling good about the work you do, how you do it, etc.
3. The ability to learn and grow on the job
4. Pay, in all its various forms, including salaries, bonus, etc.
5. Praise received for a job well done, whether from board members, fellow workers, subordinates,
the press, etc.
6. Job security
7. The opportunity to acquire and practice skills
8. The opportunity to do truly worthwhile things
12. How much of an impact does each category of motivator have on you, personally, in your
job? Please answer on the seven-point scale shown.
13. How strong is the impact of each category of motivator on your peers and/or direct reports
“back home”?
For the MBA students, the questions were similar, although (again)
referencing their last job before coming to the MBA program and, in
Question 13, asking only about their peers at that job.
Negligible
9.7% 0.0% 0.0% 1.0% 1.9% 10.6% 0.5% 1.4%
impact
Limited
21.7% 1.0% 0.0% 4.3% 4.8% 23.2% 2.9% 3.4%
impact
Moderate
37.2% 7.7% 7.7% 27.5% 22.2% 37.2% 22.2% 16.9%
impact
Substantial
19.8% 39.6% 36.2% 37.7% 37.2% 18.4% 43.0% 32.9%
impact
Powerful
9.2% 48.3% 49.8% 28.5% 30.9% 8.7% 28.5% 40.1%
impact
I live and
0.0% 3.4% 6.3% 1.0% 2.9% 0.5% 2.9% 5.3%
die . . .
Standard
1.17 0.73 0.73 0.92 1.01 1.16 0.88 1.00
deviation
Negligible
5.3% 0.0% 0.0% 0.5% 0.5% 1.0% 0.5% 0.5%
impact
Limited
15.5% 1.4% 3.4% 1.4% 1.4% 8.7% 4.3% 11.6%
impact
Moderate
33.8% 20.8% 22.2% 19.8% 13.0% 33.3% 25.6% 23.7%
impact
Powerful
18.4% 30.0% 30.0% 35.7% 31.9% 18.4% 21.3% 21.7%
impact
Standard
1.15 0.77 0.84 0.83 0.76 0.91 0.83 1.02
deviation
Table A2. Responses of MBA-Student Respondents Concerning the Eight
Categories
Negligible
17.5% 0.4% 0.0% 2.1% 0.4% 12.1% 0.0% 2.5%
impact
Limited
22.9% 1.3% 1.3% 10.0% 3.8% 27.9% 4.2% 3.8%
impact
Moderate
26.7% 11.7% 5.8% 28.3% 17.1% 31.3% 19.6% 13.8%
impact
Substantial
18.8% 37.1% 29.2% 36.7% 33.8% 18.3% 39.2% 31.7%
impact
Powerful
6.3% 44.6% 53.3% 21.7% 40.4% 6.3% 34.6% 38.3%
impact
Standard
1.35 0.83 0.79 1.07 0.94 1.21 0.89 1.18
deviation
Negligible
11.3% 0.0% 1.3% 0.8% 0.0% 3.8% 1.3% 2.1%
impact
Limited
15.8% 2.5% 3.8% 5.4% 3.8% 18.8% 7.5% 9.6%
impact
Moderate
25.8% 28.8% 22.1% 19.6% 18.8% 24.2% 27.5% 26.3%
impact
Substantial 23.3% 40.8% 43.3% 30.0% 43.3% 29.6% 42.5% 34.2%
impact
Powerful
17.9% 25.8% 27.5% 40.4% 32.9% 21.7% 20.4% 24.6%
impact
Standard
1.44 0.88 0.92 1.06 0.84 1.21 0.92 1.16
deviation
Partition these eight categories into two groups: the group of more
extrinsic motivators—consisting of “pay,” “praise,” “benefits,” and “job
security”—and the group of motivators that are more intrinsic—“feeling
good,” “learning and growing,” “acquiring and practicing skills,” and “doing
worthwhile things.” This division isn’t perfect, in the sense that (for
instance) “acquisition of skills” or “learning and growing” might be
instrumental for (later) receiving better extrinsic rewards (such as a
promotion). But, taken with an appropriate-sized grain of salt, this division
provides the following insights.
■ But when it comes to their peers and direct reports, the relative
positions of the mean scores of different categories mix between
the intrinsic and extrinsic groups.
■ And compare how the SEP participants rate each category for
themselves versus their peers and direct reports. In four cases—
the four extrinsic motivators—the mean scores are higher for their
peers than for themselves. And in the other four cases—the four
intrinsics—the mean scores are higher for themselves. For the
MBA students, again “praise” is the outlier: They rate it (praise)
as having more impact on themselves, on average, than on their
former peers.
Table A3. Average Scores for the Eight Types of Motivator
Average for Average for direct Average for Average for direct
self reports self reports
Learn &
5.55 5.04 5.66 4.98
grow
Worthwhile
5.23 4.74 5.22 4.67
things
We reprise Table 5 from the text as Table A3 here, so you can see these
three points more easily. But, as you ponder these points, please be careful
how you interpret them: These are statements about the average ratings given
to the eight categories. It is most certainly not the case that all the SEP
participants agree with these rankings.
For instance, the average score given by SEP participants for “pay” was
4.91, while “doing worthwhile things” averaged 5.23. As averages go, the
difference is statistically significant: Resorting to statistical jargon, a paired-
sample test of difference of means of the two sets of scores gives a one-
tailed critical p-value of 0.00109. This means: If the population of
executives (of the sort who choose to attend SEP) had equal average scores
for these two, the chances that a random sample of 207 (the sample we have,
to the extent that it is a random sample) would show a difference in means
this large is around 1 in 1,000. (Since this is a one-sided critical p, I should
say, “this large and in this direction.”) In the realm of social science
research, it is typical to regard any difference that gives a critical p-value
smaller than 0.05 as statistically significant. So this difference is, by usual
standards, very significant.*
Nonetheless in our sample of 207 executives, 55 (or 26.6%) gave a
higher score to “pay” than to “worthwhile activities,” and another 59
(28.5%) gave them equal scores. Even though the average score of
“worthwhile things” was very significantly greater than that of pay, over
half the respondents scored pay as having equal or greater impact on them
than “worthwhile things.”
Or to take the most extreme example possible: Compute two numbers for
each SEP respondent: the sum of the four scores he/she gives to the four
intrinsics, and the sum of the four scores he/she gives to the four extrinsics.
The average over the 207 SEP respondents of the first sum of four is 21.28.
The average for the second is 17.69. A difference-of-means test comparing
these two sets of scores (paired-sample) gives a critical p-value (one-sided)
of 1.56 × 10-35. (This is as computed by Excel and should not be taken
seriously. But the critical p-value is really, really, really small.) And yet, 21
respondents out of the 207 gave a higher sum of four to the extrinsics than to
the intrinsics, with a further 15 giving equal sums. The moral here is that
there is a lot of disagreement within the sample of respondents on the relative
impact of these categories.
This is not a case in which the “majority” or average opinion is correct
and those who hold contrary opinions are wrong. The survey asks How
strong are these motivators on you? and . . . on your peers and direct
reports? Even if the majority of people prefer apples to oranges, that doesn’t
make someone who prefers oranges wrong in any sense. And, as I’ll later
explain, there are good reasons to believe that, if (say) pay is a strong
motivator for a particular respondent, it is more likely to be a strong
motivator for his or her peers and direct reports. We can conclude: There is
no best way to motivate, in the sense that it is best for everyone, in every
situation and/or in every location/industry/ function/organization. You may
not have needed this survey to come to that conclusion. But the data are
certainly supportive.
But, granting that this is true, it becomes legitimate to ask: What (if
anything) can be learned from the average scores? I think the most
important lesson emerges if we first address the question, What accounts for
the difference in the average scores the participants gave the eight
categories for themselves versus for their peers and direct reports? Let me
remind you what we saw in this regard in the data:
■ In four cases—the four extrinsic motivators—the mean scores
given by SEP participants are higher for their peers than for
themselves. And in the other four cases—the four intrinsics—the
mean scores are higher for themselves. For the MBA students, this
pattern holds except for “praise,” which the MBA respondents
rate (on average) as having more impact on themselves than on
their former peers.
How significant are these differences? See Table A4. First for the SEP
participants and then for the MBA students, we give (paired-sample, one-
tailed) critical p-values for the differences we see between average scores
for self and then for peers/direct reports. Read this table together with Table
A3: The SEP respondents average 3.90 for themselves for “benefits” and
4.27 for their peers and direct reports, and the critical p-value for this
difference in means is 2.64 × 10-13. (The table uses scientific notation:
2.64E-13 means 2.64 × 10-13.) The p-values are all extraordinary small; the
only p-value that is even close to the edge of (non)significance is “praise”
for the MBA students, with a critical p of 0.0108. And remember, praise is
the one “anomalous” category, in the sense that the SEP respondents gave a
higher score to “praise” for their peers/direct reports than to themselves,
while the MBA students on average scored “praise” higher for themselves.
2. Their opinions about themselves are largely correct, but they are
wrong about their peers and direct reports, who (in truth) are
motivated similarly to how the respondents themselves are
motivated.
The SEP participants, confronted with these data, have been split over which
of these four possible truths is in fact the truth.
4. As for adherents to Possible Truth #4, they say there is merit in all
these arguments, and the true state of affairs lies somewhere in the
middle.
I have been assuming that the division into “four intrinsics” and “four
extrinsics” makes sense. Certainly, this division is supported by the data in
Tables A3 and A4 and, in particular (and with the exception of “praise” for
MBA students), the reversal of average impact between self and others that
we see in surveys, in literally all cases of which I know.
But is there more than this? In particular, are individuals either “more
intrinsically motivated” or “more extrinsically” motivated? Can we divide
respondents into one of two buckets of this sort? Evidence in favor (or not)
of this hypothesis can be gathered by looking at the correlations in scores
between pairs of the eight categories. See Table A5. This provides, first for
the SEP participants and then the MBA students, the cross-category
correlations on the scores they gave for themselves.
The eight categories are ordered so that the four extrinsics are first, and
then the four intrinsics. The hypothesis that individuals are either “intrinsic”
or “extrinsic” in orientation would be supported if we saw larger
correlations in the upper-left and lower-right “triangles” and smaller or even
negative correlations in the upper-right square. And we certainly see a
tendency in that direction, although there are some anomalies.‡
Table A5. Correlations in Self-Impact Scores of the Eight Types of
Motivator
Learning &
0.644 0.242
growing
Skills 0.376
Learning &
0.535 0.286
growing
Skills 0.304
There is more to say about the eight-category portion of the survey but,
first, we should catch up on the five-channel portion.
Five Motivational Channels
In the second part of the survey, respondents were asked about five
motivational channels: personal, tangible rewards; personal, intangible
rewards; work that is interesting or exciting; work leading to success of the
organization or workgroup; and work that is socially important. For both
themselves and others, they were asked to assess the effectiveness of each
channel and to say which channel was most descriptive in eliciting best work
or consummate effort where, for SEP participants, the “others” were their
direct reports and, for the MBA students, it was their peers on their last job.
Results are provided in Tables 2 and 4 in Chapter 3 (on pages 62 and 65),
and repeated in Chapter 8. (The means and standard deviations reported are
computed based on the numerical scale in which not at all effective = 1, of
limited effectiveness = 2, and so forth.)
In terms of motivating others, respondents were asked Question 9,
concerning how similar their direct reports or peers were in terms of
motivation. The distributions of answers to this question for the two groups
are provided in Table A6.
Just as in the other part of the survey, there is a lot going on in these data.
Some highlights are:
Direct reports are quite different. To motivate their best work, I must vary the techniques
58.9%
used:
Direct reports are different, but they can be (and are) motivated in largely similar ways: 35.7%
Direct reports are very similar in how to motivate them, and I treat them similarly,
5.3%
accordingly:
Peers were quite different. To motivate their best work, different techniques were
29.2%
necessary
Peers were different, but they could be motivated in largely similar ways: 51.7%
Peers were very similar in terms of motivation, and were treated similarly, accordingly: 19.2%
■ This variety in answers is buttressed by the fact that, for the SEP
respondents, nearly 60% say that their direct reports are so
different that motivational techniques must be tailored to the
individuals. Contrast this with the 70% of MBA students who say
that similar motivational techniques work for their peers, even
though they perceive substantial difference in what motivates
those peers.
Please do not overprocess these numbers. The sizes of the various groups are
such that statistical significance is rare.§ My take on what is significant in
these numbers—and I mean significant in terms of the magnitude of the
effects as much as their statistical significance—are: (a) None of these
categories explains a lot. People in different demographic groups give
responses that are more alike, in general, than they are different. (b) People
in the public sector, however, are a lot more motivated by social contribution
and by benefits than are other groups. (c) Chairs/CEOs and, to a somewhat
lesser extent, COOs, give higher scores to themselves in terms of motivation
by the success of their organization. (d) People in the financial sector give
themselves high average scores for pay in the eight-category portion of the
survey; not so much in the five channels. But when it comes to their
peers/direct reports, they certainly believe the stereotypes. And the
consulting types are not far behind (especially when you consider that the
consulting group mixes in at least a few academic administrators who went
through the SEP, and I have a hard time thinking that academics would admit
to being highly motivated by pay).
Table A9. SEP Average Scores by Demographic Groups
Table A10. MBA Students’ Average Scores by Demographic Groups
Concerning the MBA students, similar data are presented in Table A10.
Make of these what you will.
While it is difficult to find much explanatory power from demographic
groups within each of the two surveys, there are marked differences between
the responses given by MBA students and SEP participants. The
demographic differences between the two populations are manifest: SEP
participants are older and, in virtually all cases, highly accomplished senior
managers. Stanford MBA students are considerably younger, and while they
already have a record of considerable success in their careers, those careers
are (so far) a good deal shorter. It perhaps isn’t accurate to say that the SEP
participants are what the MBA students aspire to become; Stanford MBA
students are, overall, less interested in working for established companies
and much more interested in starting their own, which may also account for
some of the differences. In any event, Table A11 directly compares the
average scores given by SEP participants and MBA students, with critical p-
values for difference-of-means tests.¶
The large (and statistically significant) differences are obvious. MBA
students are more motivated (on average) by praise. They care less about
benefits. And the biggest difference: They are much less motivated, and (if
we believe the survey results) their peers are much less motivated, by
success of the organization for which they work. (It stands to reason that the
MBA students have less career concerns involving their former employer
than do the SEP participants, involving their current employer. So this
comparison may be a bit unfair as a test of MBA-student loyalty.)
One inexplicable result in these data concerns the motivational power of
pay. In the five-channel portion of the survey, MBA students gave pay a very
slightly higher average score than did the SEP participants for themselves,
but a significantly higher score for the peers than the SEP participants gave
for their direct reports. But in the eight-category portion, the SEP participants
gave higher scores to pay than did the MBA students, both for themselves and
for their “others.” And the significant difference in this part of the survey was
in self-scores. The moral of this anomaly may be: Most such survey results
should be regarded with at least some skepticism.
Table A11. Mean-Score Comparisons: SEP Respondents versus MBA-
Student Respondents
Five channels
3.1E-
Pay, etc. 3.85 3.88 0.37 3.77 4.06
04
4.1E- 3.2E-
Praise, etc. 4.15 4.43 4.06 4.25
04 03
5.1E-
Social good 3.78 3.60 3.32 3.14 0.037
02
Eight categories
4.7E-
Benefits 3.90 3.50 4.27 4.08 0.012
04
2.0E-
Feeling good 5.45 5.39 5.08 4.94 0.038
01
1.2E-
Pay 4.91 4.62 5.14 5.10 0.312
03
3.9E-
Praise 4.99 5.24 5.17 5.09 0.142
03
Feeling good 0.36 0.54 0.19 0.16 0.16 0.13 0.23 0.12
Learn & grow 0.16 0.23 0.41 0.08 0.05 0.19 0.38 0.14
Job security 0.22 0.07 0.08 0.20 0.12 0.46 0.15 0.11
Acquire skills 0.17 0.22 0.31 0.05 0.05 0.26 0.51 0.27
Worthwhile
0.17 0.14 0.14 -0.12 -0.02 0.17 0.27 0.63
things
Pay, etc. Praise, etc. Interesting work Org. success Social good
Feeling good 0.17 0.54 0.25 0.01 0.11 0.14 0.13 0.26
Learn & grow 0.19 0.24 0.33 0.14 0.17 0.09 0.27 0.14
Job Security 0.25 0.16 0.16 0.25 0.25 0.43 0.21 0.20
Worthwhile
0.13 0.33 0.19 -0.11 0.02 0.17 0.15 0.60
things
Pay most
descriptive for 17.4% 40.6% 26.1% 7.5% 12.3% 14.3%
other
Praise is most
20.2% 18.8% 65.2% 7.5% 15.8% 19.0%
descriptive
Interesting
work is most 26.6% 25.0% 8.7% 58.5% 21.1% 4.8%
descriptive
Organizational
success is most 23.7% 6.3% 0.0% 20.8% 49.1% 19.0%
descriptive
Social good is
most 12.1% 9.4% 0.0% 5.7% 1.8% 42.9%
descriptive
Pay most
descriptive for 28.8% 65.7% 34.8% 18.7% 15.8% 7.7%
other
Praise is most
22.1% 17.1% 39.1% 11.0% 31.6% 15.4%
descriptive
Interesting
work is most 34.2% 17.1% 17.4% 58.2% 36.8% 15.4%
descriptive
Organizational
success is most 4.6% 0.0% 1.4% 5.5% 15.8% 7.7%
descriptive
■ Beyond this, no one size fits all. When it comes to the power and
impact of different motivational techniques other than interesting
and/or exciting work, there is no strong consensus. Different
managers are motivated differentially by the different techniques.
_____________
*
The smallest difference for the SEP participants, for themselves, is between praise, at 4.99, and
acquiring skills, at 5.05. This difference is not statistically significant; the critical p-values (paired-
sample, one-sided) is 0.26. Next smallest is pay versus acquiring skills; the critical p-value for this
comparison is 0.062.
†
The title of Chip Heath’s original article on the subject, “Lay Theories of Motivation Overemphasize
Extrinsic Incentives,” clearly indicates that he holds this view.
‡
In both populations, job security has “high” correlation with skills. Perhaps respondents think that
building their skills enhances their job security. And for the MBA students, “feeling good” and “praise”
have relatively high correlation. Perhaps MBA students look for praise from external sources because
they need external validation for how they are doing.
It is worth observing, perhaps, that all the correlations tend to be positive, if not hugely so. This
reflects a “scaling” individual-fixed effect: Some people are more likely to use the right-hand scale
options; others the left-hand scale options.
§
If you want to engage in testing statistical significance for any specific hypotheses you might have, and
if you know how, be my guest. It is probably reasonable to run such tests with an assumed equal
standard deviation of 1 for each underlying population.
¶
These tests are, of course, not paired-sample tests. I do use implied equal variance, since the standard
deviations are quite similar.
NOTES
I would like to acknowledge and thank a number of people, both groups and
individuals, for their contributions to this book.
James Baron is first and foremost. It would not be at all inaccurate to
describe this book as the “Classics Illustrated” version of the textbook,
Strategic Human Resources: Frameworks for General Managers, that I
wrote with Jim. People are complex animals, so managing them is complex,
and if the reader finds the content of this book to be of interest and of use, I
cannot too strongly recommend the longer, more detailed, and more nuanced
textbook. But, beyond this, Jim taught me a lot of whatever wisdom this book
contains, and I’m grateful to him for that and for our friendship.
Other colleagues, both economists but especially members of Stanford
Graduate School of Business (Stanford GSB) faculty who are not
economists, have been generous with their time and insights as I put this book
together. In alphabetical order, I am grateful in particular to Jennifer Aaker,
William Barnett, Glenn Carroll, Francis Flynn, Robert Gibbons, Deborah
Gruenfeld, Michael Hannan, Tamar Kreps, David Larcker, Edward Lazear,
Neil Malhotra, James March, William Meehan, Dale Miller, Benoit Monin,
Charles O’Reilly, Paul Oyer, Jeffrey Pfeffer, Hayagreeva Rao, Peter Reiss,
Kathryn Shaw, Jesper Sorensen, Larissa Tiedens, and Mark Wolfson. (I
apologize for any inadvertent omissions.)
This book is, of course, about management, and I learned a lot about
management in general, managing human resources, and motivation, by
working for and with a master of these crafts, Robert Joss.
Generations of MBA students and Stanford Executive Program
participants learned this material along with me and, in so doing, taught me a
lot.
Many people at W. W. Norton contributed to the construction of this book,
perhaps too many to cite by name. But, in particular, Jeff Shreve picked up
this project in midstream and made many important contributions to the
organization and exposition of the book. And when Jeff left Norton near the
completion of the book, Brendan Curry and Nathaniel Dennett, together with
production manager Julia Druskin, did an excellent job in shepherding the
book to the end.
Finally, I was encouraged to begin and, every step along the way, to
continue working on this book by Jack Repcheck, who previously had edited
two textbooks on economics that I wrote. Jack’s untimely death prevented
him from seeing this book to completion, but both as friend and editor, his
“imprint” is found everywhere within these pages. It is with profound
sadness for his passing but with wonderful memories of him that I dedicate
this book to his memory.
INDEX
Page numbers listed correspond to the print edition of this book. You can use your device’s search
function to locate particular terms in the text.
categories of employees
consistency of HRM practices across, 205
demographic, discrimination based on, 195–97
clarity, expectancy theory and, 132–33
Clifford, Joyce, 127
cognitive psychology
cognitive biases from, 121
consistency of HRM practices and, 203–4
collaboration, 179–80
Collaborative Intelligence (J. Richard Hackman), 181
compensation; see also stock options
at Artisans’ Alliance, 3–6, 159–62
overestimating importance of, 116–18
for patient hospitalizations, 158
performance-based, see pay for performance
with stock options, 50–57
tournament-style, 33–34
competence, self-determination theory and, 144, 145
confirmation bias, 121
connections to job
expressive vs. instrumental, 108–9
types of organizations and, 109–12
consistency of HRM policies and practices, 202–7
across categories of employees, 205
over time, 205–7
psychology of perception and cognition and, 203–4
recruitment and, 204
technology and, 202–3
consummate performance, 6
contract(s)
psychological, 111, 201
relational, 99–100
contributions, equity theory and, 139–43
cooperation, reciprocal, 174–76
credibility, 81–82
cultures, see organizational cultures
IBM, 205
implicit contracts, 99–100
improving motivation, 183–209
business strategy and, 184–87
economic environment and, 187–90
employee characteristics and, 195–200
employment relationships and, 200–201
HRM policies and practices and, 202–7
job designs and, 194–95
legal environment and, 187–90
organizational culture and, 208–9
pool of job applicants and, 201–2
social environment and, 187–90
work technology and, 190–94
incentive stock options, 53
incentive theory, 14; see also pay for performance
conceptual biases created by, 89–90
economics of employment relationships and, 88–92
fundamental trade-off in, 23–26
governance in, 91
stock options and, 50–57
independent contractors, 97
informal agreements, 100
instrumental connections to job and work, 108
instrumentality, in expectancy theory, 132–34
“insurance,” with pay for performance, 30–31
interdependence of employees’ work, 34, 191
interesting work, 210–14
intrinsic motivation, 144
reasons for ratings of, 234–40
in self-determination theory, 144–46
survey results for, 228–34
undermining and, 150–56
Uber, 98
undermining, 150–56, 164
Unilever, 111–12
For information about permission to reproduce selections from this book, write to Permissions, W. W.
Norton & Company, Inc., 500 Fifth Avenue, New York, NY 10110
ISBN: 978-0-393-25409-9
W. W. Norton & Company, Inc., 500 Fifth Avenue, New York, N.Y. 10110
www.wwnorton.com