Quality
and Moral Hazard: When Your Organization’s Insulation From Consequences
Creates the Risks It Was Trying to Avoid
The warranty claim landed on a Tuesday morning. A medical device
company had shipped 40,000 units with a sealing defect that should have
been caught at three different inspection points. The cost of the recall
was staggering — $12 million and counting. But what made the quality
director’s blood run cold wasn’t the number. It was the realization that
every single person involved in the process had done exactly what they
were incentivized to do.
The production manager had hit his throughput targets. The inspection
team had met their cycle time goals. The supplier quality engineer had
approved the incoming material because the Certificate of Conformance
looked fine. The engineering team had signed off on the process
validation because all the paperwork was complete. Nobody did anything
wrong — according to the metrics they were measured on.
And that’s how 40,000 defective units ended up in hospitals.
This is moral hazard in quality management. It happens when the
people making decisions don’t bear the full consequences of those
decisions. It happens when safety nets designed to protect against
failure actually encourage the behavior that causes failure. It happens
in every organization, every day, and most quality professionals don’t
even recognize it.
What Moral Hazard Actually
Means
The concept comes from insurance economics. When you buy
comprehensive car insurance, you might park in riskier neighborhoods or
drive a little faster because you know the insurance company will cover
the damage. The protection meant to shield you from loss subtly changes
your behavior in ways that make loss more likely.
That’s moral hazard. The insulation from consequences doesn’t just
reduce the pain of failure — it changes the calculus of risk itself.
In quality management, moral hazard operates the same way, but the
mechanisms are organizational, not contractual. And the consequences are
measured not in insurance claims but in defective products, lost
customers, and sometimes — in industries like automotive, aerospace, and
medical devices — in human lives.
The Five Faces of
Moral Hazard in Quality
1. The Inspection Safety Net
Here’s a pattern I’ve seen in dozens of manufacturing organizations:
the production team knows there’s a final inspection at the end of the
line, so they invest less energy in getting it right the first time. Why
would they? Someone else will catch it.
I worked with an automotive parts supplier where the defect rate at
final inspection was running at 4.2%. When we dug into the data, we
found that operators at the upstream welding station had stopped
performing their in-process checks — because quality always caught the
bad ones downstream. The inspection that was supposed to be a safety net
had become a crutch.
The math is brutal. When upstream processes rely on downstream
inspections, the cost of quality multiplies at every step. A defect that
costs $0.50 to fix at the welding station costs $5 at assembly, $50 at
final inspection, and $5,000 if it reaches the customer. The safety net
doesn’t just enable sloppy work — it amplifies its cost by orders of
magnitude.
2. The Warranty Shield
In industries with extended warranties or service contracts, the
production organization knows that field failures are someone else’s
problem. The warranty department will handle it. The service team will
replace it. The customer service representatives will apologize for
it.
I consulted with a consumer electronics manufacturer whose return
rate had been climbing steadily for three years. The quality team had
identified the root cause — a thermal cycling issue in the power supply
— but the production team resisted the fix because it would slow down
the line by seven seconds per unit. Their bonus was tied to units per
hour, not to warranty costs. The warranty budget came from a different
cost center entirely.
The organization had created a perfect moral hazard: the people who
could prevent the defect weren’t the people who paid for it. The seven
seconds they saved on each unit cost the company $8 million in warranty
claims that year.
3. The Approval Committee
Diffusion
Many organizations route critical quality decisions through
committees — change control boards, material review boards, design
review panels. The intention is good: collective wisdom prevents
individual mistakes. The side effect is catastrophic.
When twelve people sign off on a decision, no single person feels
responsible for it. The engineering director assumes the quality manager
checked the specs. The quality manager assumes the production supervisor
validated the process. The production supervisor assumes the supplier
quality engineer verified the material. Everyone approved it. Nobody
owned it.
I saw this at a pharmaceutical company where a change in a tablet
coating supplier sailed through the change control board with unanimous
approval. Six months later, the dissolution rate was out of
specification. In the investigation that followed, every board member
pointed to another member as the one who was supposed to have caught it.
The committee that was designed to prevent failures had distributed
responsibility so thinly that nobody felt the weight of it.
4. The Audit Armor
Here’s one that catches even experienced quality professionals off
guard: the audit itself can become a source of moral hazard.
When an organization knows it will be audited — by a customer, a
registrar, a regulatory body — it often directs enormous energy toward
audit readiness. Procedures are updated. Training records are completed.
Calibration stickers are checked. The facility is cleaned. The auditors
arrive, see a well-organized quality system, and issue a clean
report.
Then the auditors leave, and everything gradually relaxes back to its
real state.
The audit, which was supposed to be a genuine check on quality system
effectiveness, becomes a performance. The moral hazard is subtle: the
organization isn’t investing in quality because it believes in quality.
It’s investing in the appearance of quality because it knows someone
will be checking. And the energy that goes into the performance is
energy that doesn’t go into the actual practice.
The most revealing metric I’ve ever tracked isn’t the audit score.
It’s the delta between audit-period performance and the thirty days
after the auditors leave. In organizations with strong moral hazard,
that delta is enormous.
5. The Blame Deflection
Perhaps the most dangerous form of moral hazard is the one built into
how organizations handle failures.
When a defect escapes, most organizations launch an investigation.
They find the person who made the error. They retrain them. They add an
inspection. They close the corrective action. The system records that
the problem has been solved.
But here’s what really happened: the organization took a systemic
failure and reduced it to an individual failure. The operator who missed
the defect becomes the cause. The retraining becomes the corrective
action. And the system that created the conditions for the error — the
inadequate lighting, the unclear work instructions, the impossible cycle
time, the designed-in ambiguity — remains untouched.
This is moral hazard at the organizational design level. The system
is insulated from the consequences of its own design failures because
individuals serve as the shock absorbers. As long as you can blame a
person, you never have to fix the process.
Why Moral Hazard Is So Hard
to See
Moral hazard is invisible by design. The people caught in it aren’t
acting maliciously — they’re acting rationally within the incentive
structure they’ve been given. The production manager optimizing for
throughput isn’t trying to ship defects. The operator skipping an
in-process check isn’t trying to sabotage quality. The committee member
rubber-stamping a change isn’t trying to approve something
dangerous.
They’re all responding to the signals the organization sends about
what it actually values. And those signals — metrics, incentives,
structures, processes — are often in direct conflict with the stated
quality objectives.
The organization says “quality first” but measures throughput first.
It says “zero defects” but rewards volume. It says “continuous
improvement” but punishes the line stoppages that come from actually
stopping to fix problems. The moral hazard isn’t hidden in the culture.
It’s built into the measurement system.
The Structural Fixes
Addressing moral hazard requires structural change, not cultural
exhortation. You cannot poster your way out of an incentive problem.
Here are the interventions that actually work:
Align consequences with decisions. The people who
make quality decisions must feel the impact of those decisions. If
production controls the line speed, production must own the cost of
escapes — not just the throughput numbers. If engineering designs the
process, engineering must participate in the corrective actions when the
process fails. This doesn’t mean punishment. It means shared
accountability measured in the same currency.
Make the hidden costs visible. Most organizations
have no idea what their quality failures actually cost because the costs
are distributed across departments. Warranty costs sit in customer
service. Scrap costs sit in manufacturing. Rework costs sit in
operations. Reputation costs sit nowhere because nobody measures them.
Build a single cost-of-quality model that aggregates all of it, and make
it visible to every decision-maker.
Eliminate unnecessary safety nets. Every inspection
point should have a justification. If you have three inspections doing
essentially the same thing, you don’t have three layers of protection —
you have three layers of moral hazard, each one enabling the others to
relax. Reduce redundant inspections and invest the saved resources in
process capability instead. A capable process doesn’t need a safety
net.
Flatten approval structures. Replace committees with
clearly assigned decision authority. When someone makes a material
disposition decision, their name should be on it. Not as a punishment
mechanism, but as an ownership mechanism. People make better decisions
when they know those decisions are traceable to them. Not because they
fear blame — because they feel responsibility.
Audit the system, not the paperwork. Shift audit
focus from documentation to performance. Don’t ask whether the procedure
exists. Ask whether the procedure produces the intended result. Don’t
check whether training was completed. Check whether the trained people
can do the work correctly. The difference between compliance and
competence is the difference between theater and quality.
Measure what you actually want. This is the single
most powerful lever. If you want zero defects, stop measuring operators
on throughput alone. If you want process improvement, stop rewarding
fire-fighting heroes. If you want cross-functional collaboration, stop
evaluating departments in isolation. Every metric is an instruction.
Make sure you’re instructing what you actually want.
The Deeper Principle
Moral hazard in quality management is ultimately a question of
organizational architecture. The question isn’t “How do we make people
care more about quality?” The question is “What structures have we built
that make it rational for people to care less?”
Because here’s the uncomfortable truth: your quality system is
perfectly designed to produce exactly the results it’s producing. If you
have a 2% defect rate, your organization’s incentive structures,
measurement systems, approval processes, and accountability mechanisms
are all precisely calibrated to generate a 2% defect rate. Not because
anyone wants 2% defects. Because the system rewards behaviors that
produce 2% defects.
Changing the results means changing the system. And changing the
system means having the courage to look at the structures you’ve built
and admit that some of them — the ones you created to protect quality —
are the very things undermining it.
That medical device company didn’t fix their problem by retraining
operators or adding inspections. They fixed it by restructuring their
metrics so that the production manager’s bonus was calculated using a
formula that included warranty costs, not just throughput. Within six
months, the defect rate dropped from 4.2% to 0.3%. The operators didn’t
get better. The system did.
Moral hazard isn’t a character flaw in your people. It’s a design
flaw in your organization. And the sooner you stop looking for someone
to blame and start looking at the structure that makes the behavior
rational, the sooner your quality system will start producing the
results you actually want.
Peter Stasko is a Quality Architect with 25+ years
of experience transforming organizations across automotive, aerospace,
and pharmaceutical industries. He specializes in helping companies see
the structural patterns their quality systems create — and redesigning
those systems so that doing the right thing becomes the easiest thing to
do.
