Library · Long read

When the systems we build become the barriers we defend

The long version of our carousel. Eleven research reports, fourteen documented cases and one question for anyone who runs, funds or works inside an organization: does your revenue depend on the problem staying around?

Rigby Consulting & FirmNovember 2026About 30 minutes47 sources

Average years a company stayed in the S&P 500

33Measured, 1965
21Measured, 2020
14Innosight’s forecast for 2026
The drop is real. The exact number keeps moving: each forecast since 2016 has landed between 12 and 20 years. The figure is a rolling seven-year average of time on the index, and companies also leave through mergers, not only through failure.1,2

Protecting the problem doesn’t protect the company.

In short

  1. A fix becomes a barrier under three conditions. The people being judged control the measure. The people who pay can’t leave or be heard. Nobody outside can force a correction.
  2. It rarely starts as a plan. In most cases we studied, no record shows anyone set out to protect the problem. The arrangement drifted until protecting it became part of the job. Where a court, a plea or an admission says it was deliberate, we say so.
  3. Insiders already know. In a survey of 401 financial executives, 78% said they would give up real economic value to keep earnings smooth.3
  4. Some barriers are the part that still works. The FDA reviewer who refused to approve thalidomide on thin evidence was a barrier too.26
  5. What changes outcomes is structural. Change who measures, who decides or what triggers payment. Disclosure and good intentions on their own move very little.

1 · The question

Does your revenue depend on the problem staying around?

Every business earns its money from a problem. A plumber’s revenue depends on leaks. A security firm’s depends on attackers. A hospital’s depends on illness. None of that is a scandal. It’s what a business is.

The question we’re asking is narrower, and harder to sit with. It isn’t whether your revenue depends on a problem. It’s whether it depends on the problem staying. A plumber who fixes the leak gets paid. A plumber who’s paid for every visit, and who also writes the inspection report, has a different set of incentives, even if he never acts on them.

Our founder put the idea in one sentence that started this whole series:

“When you try to take relevance away from the problem, you build barriers as a business.”Devon Rigby, Rigby Consulting & Firm

How a fix turns into a barrier

Across the research, the same three-stage arc kept showing up.

  • Purpose. A system is built to solve something real: house military families, keep the lights on, buy parts for aircraft, get people covered by insurance.
  • Then interests. Budgets, headcount, careers, bonuses and contracts grow around the system. Each one is reasonable on its own. Together they form a constituency for the system as it is.
  • Then the arrangement gets defended. At some point, keeping the arrangement intact becomes more valuable to the people inside it than finishing the job it was built to do. Nobody has to decide this out loud. The measures, the rewards and the reporting lines decide it for them.

Every company runs two clocks

The easiest way we’ve found to see the tension is to picture two clocks running in the same building.

The business clock counts the quarter, the number and the money back. It’s loud, it’s precise and it gets reviewed every few weeks. The development clock counts the problem, the people and the future. It’s quiet, it’s hard to measure and it rarely has its own line on the dashboard.

Most of the time the two agree. Solving the customer’s problem well is how you hit the number. The trouble starts on the days they disagree: the maintenance that would hurt this quarter, the project that pays off after the bonus period, the fix that would shrink a contract. On those days the business clock usually wins, because it’s the one being watched.

Why the S&P 500 number belongs here

The carousel opened with a number from Innosight’s corporate longevity research. In 1965, a company stayed in the S&P 500 for an average of 33 years. By 2020, the measured figure was 21. Innosight’s forecast for 2026 was 14.1,2

We show both the measured figures and the forecast because the difference matters. The drop from 33 to 21 is measured. The 14 is a projection, and the projections have moved: each forecast since 2016 has landed somewhere between 12 and 20 years. The figure is also a tenure average on an index, not a lifespan. Companies leave the S&P 500 through mergers and acquisitions as well as decline.

So here’s what the number does and doesn’t tell you. It doesn’t prove that protecting problems shortens a company’s life. Innosight’s data can’t show that, and we won’t claim it. What it shows is that the window in which any arrangement stays profitable has been getting shorter. Protecting the problem doesn’t protect the company is our reading of that, not a finding in the data. The rest of this report is the evidence for why we read it that way.

2 · The pattern

Three conditions, in almost every case

We looked at military housing, defense parts, banking, energy, autos, healthcare, public programs and technology. The industries had little in common. The structure did.

1

The people judged control the measure

The party being evaluated produces the records it’s evaluated on: the work orders, the status reports, the cost data.

2

The people who pay can’t leave or be heard

Military families can’t pick another landlord on base. Ratepayers can’t pick another grid. Frontline staff can’t safely report up.

3

Nobody outside can force a fix

There may be oversight, but no independent party with both the information and the binding authority to make a correction stick.

None of the three proves wrongdoing on its own. Plenty of honest organizations produce their own performance data, and plenty of customers have few alternatives. But where all three line up, the system can keep reporting success while the problem it was built to solve stays in place. That’s the condition we mean when we say a system has become a barrier.

A related finding from the synthesis report surprised us. Control of verification may matter as much as control of assets. You don’t need to own every resource to protect an arrangement. You only need to decide what counts as success. We treat that as a hypothesis the cases support, not as a measured ranking.

How much of it is on purpose?

This is the question people ask first, and it’s the one the evidence answers least often. We built a six-rung ladder to keep ourselves honest. Each rung is a different explanation for why a problem stays around.

  1. 01Couldn’t avoid it. The constraint was real: money, physics, law or time.
  2. 02Nobody owned it. The problem fell between teams, budgets or agencies.
  3. 03Nobody checked. The measure looked fine, so no one looked past it.
  4. 04Chose not to look. Signals were there, and attention went elsewhere.
  5. 05Paid to leave it. The rewards made leaving the problem in place the rational move.
  6. 06Blocked the fix. Someone with authority acted to stop a correction.

Our rule is simple. We name a rung only where a record names it. A guilty plea, a court finding or a formal admission can put a case on rung five or six. A bad outcome on its own can’t. Most of what we studied sits on rungs two through four: people who stopped asking, inside structures that never made them ask. That isn’t an excuse. It’s a more useful diagnosis, because rungs two through four are the ones a leader can actually redesign.

3 · What it looks like

The cases, grouped by mechanism

Each case below comes with two lists: what the record proves, and what it doesn’t. The second list matters as much as the first. A report that overstates its evidence becomes one more system defending its own numbers. Every case is described as the cited public record describes it: pleas, court orders, audits, agency reports and peer-reviewed or working-paper research. We don’t allege anything beyond what those records state.

How performance pay gets verified

Military housing · award fees

Balfour Beatty Communities ran privatized housing on military bases under agreements that paid performance incentive fees. In December 2021, according to its plea agreement with the Justice Department, the company pleaded guilty to a fraud scheme in which maintenance records were falsified, from 2013 to 2019, to obtain those fees. It agreed to pay about $65.5 million in criminal fines and restitution.8 The plea covers that company and that conduct only.

The pattern is older than one company. In 2005, GAO found the Defense Department had paid an estimated $8 billion in award fees on contracts regardless of outcomes.11 A later DoD review, reported by GAO, found contractors were paid a median of 93% of available award fee.12

Proven

A defined scheme of falsified performance records, admitted in a plea. Award fees that, by GAO’s measure, didn’t track outcomes.

Not proven

That all housing contractors, or all incentive pay, works this way. Performance pay often improves delivery when someone else verifies the result.

How a price gets judged without cost data

Defense spare parts

In 2019 the DoD Inspector General reviewed spare parts bought from TransDigm. It found $16.1 million in excess profit on 46 parts sold for $26.2 million between 2015 and 2017, with margins on individual parts ranging from 17% to 4,451%.13 The IG found that contracting officers had followed the acquisition rules, but lacked the cost information needed to judge whether the prices were reasonable. After a House hearing that May, TransDigm agreed to voluntarily refund the $16.1 million. A second IG audit, released in December 2021, identified at least $20.8 million in excess profit on later purchases, measured the same way.14 Neither audit was a finding that TransDigm broke the law.

Proven

Profit far above the IG’s benchmark, on purchases made through a lawful process, where the buyer couldn’t see the seller’s costs.

Not proven

That anything was illegal. The IG did not say it was. “Excess” means above the IG’s benchmark, which isn’t a legal ceiling, and sole-source production can justify higher returns.

How a known fix gets adopted, or doesn’t

The Texas grid · PG&E

After a 2011 cold snap knocked out generation in Texas, federal regulators recommended winterization. By the regulators’ own later account, those recommendations largely stayed voluntary. In February 2021 a far worse freeze hit. FERC and NERC found that 81% of freeze-related generator outages happened at temperatures above the units’ own stated design limits.15 The Texas Department of State Health Services counted 246 deaths related to the storm, from all causes, not only power loss.16

PG&E shows the opposite failure: the oversight was mandatory and still fell short. After the San Bruno pipeline explosion, the utility spent five years on federal criminal probation. In his order ending that probation, the supervising judge wrote that while on probation PG&E had set at least 31 wildfires that killed 113 Californians, a period that includes the 2018 Camp Fire, for which the company pleaded guilty to 84 counts of involuntary manslaughter. He wrote that he “must acknowledge failure.”17

Proven

Federal regulators’ account of a fix left largely optional for a decade. A federal judge’s own written assessment of a probation he supervised.

Not proven

That 2011 rules would have prevented 2021. The 2021 storm was more severe, and 246 is the state’s count; other estimates run higher. The 113 deaths overlap other counts and must never be added on top of them. Texas made winterization mandatory afterward.

How oversight powers get used

Military housing oversight

Private companies own and operate about 99% of family housing on U.S. military installations. In 2023, GAO reported that DoD can place companies on improvement plans, withhold incentive fees and replace project partners. It added that DoD “has never identified the need” to fully terminate a project.10

Proven

A termination power that, as of GAO’s review, had never been used.

Not proven

That it should have been. These projects are long ground-lease partnerships, not ordinary contracts, and officials told GAO the structure lets them fix performance without collapsing a whole project.

How information moves inside a large organization

GM · Wells Fargo

In 2015 General Motors entered a deferred prosecution agreement with federal prosecutors. In it, according to the Justice Department, GM admitted that from 2012 to 2014 it failed to disclose a deadly ignition-switch defect and misled consumers about it.18 Our reading of the public record is that the earlier years look more like a defect lost between engineering, legal and management channels. The 2012 to 2014 conduct is what GM admitted. Both can be true in the same company.

At Wells Fargo, after the 2016 sales-practices settlements, the board’s independent directors commissioned their own investigation into how the problem developed and how information about it reached the top, and published it in 2017.19 In 2018 the Federal Reserve capped the bank’s growth until it improved its governance and oversight. The Fed lifted the cap in June 2025 and ended the broader enforcement action in March 2026.20,21

Proven

GM’s admitted nondisclosure, 2012 to 2014, as stated by the Justice Department. The Fed’s actions and dates, as stated by the Fed.

Not proven

That any individual GM executive knew, or that the whole decade was a cover-up. A deferred prosecution agreement is not a conviction. That lifting the cap means anything beyond what the Fed determined. A regulator’s sign-off is a milestone, not a warranty.

How a finished process differs from a solved problem

Medicaid · year-end spending

When pandemic-era continuous Medicaid coverage ended, states had to renew everyone. In the reporting window MACPAC analyzed, 68.7% of terminations were procedural: paperwork wasn’t completed or returned, rather than a finding that the person was ineligible.22 The process was completed. Whether the right people kept coverage is a separate question the process didn’t answer.

Budgets show the same habit. A study of federal contracts found spending in the last week of the fiscal year ran roughly five times the weekly average, and year-end IT projects were rated lower quality. Agencies allowed to carry money into the next year showed much less of the rush.37

Proven

A large share of coverage losses happened through procedure. Use-it-or-lose-it rules pull spending into the final week.

Not proven

That every procedural termination removed an eligible person, or that anyone intended it. Some people terminated on procedure were ineligible.

How quality gets judged when it’s hard to see

Nursing homes · public research · Bell

Nursing-home quality is hard for patients and families to inspect. A revised academic study of private-equity acquisitions estimated an 11% increase in mortality among the patients it studied. The estimate depends on the study’s design and doesn’t describe any particular owner.23

Public money can build the foundation for products that others own. A study of 210 new drugs approved from 2010 to 2016 found every one was linked to NIH-funded research, mostly basic science on the drug’s target.24 That’s a link, not a claim that NIH invented the drugs; private developers carried real cost and risk. And when the 1956 antitrust settlement forced Bell to license its patents, follow-on innovation rose outside telecommunications. Inside telecom, where Bell still controlled the market, it didn’t show the same gain.25 Opening one door didn’t open every door.

Proven

Qualified estimates from published research designs, each with its own limits.

Not proven

A verdict on all private equity, on drug pricing or on any owner’s intent. These cases show where the three conditions show up, not who meant what.

4 · Why good people defend it

The admissions come from inside

The most persuasive evidence in this series didn’t come from critics. It came from the people running the systems, describing what they’d do.

78%said they’d give up real economic value to keep earnings smooth
55%said they’d avoid starting a project that creates value if it meant missing the quarter’s earnings target
80%said they’d cut discretionary spending, maintenance included, to hit a target

Graham, Harvey & Rajgopal, survey of 401 financial executives, Journal of Accounting & Economics, 2005.3

Read those numbers again with the two clocks in mind. These weren’t people defending fraud. They were describing ordinary pressure, and the ordinary way the business clock wins. The authors noted that the executives didn’t like the system either; they felt they had to play by it or risk their jobs. Nobody in that survey would call a delayed project a barrier. They’d call it hitting the number. That’s how the future gets shelved without anyone deciding to shelve it.

A follow-up study asked 169 CFOs about other companies. They estimated that about 20% of firms manage earnings to misrepresent performance.4 These are survey answers about what people say they would do or what they believe others do. They aren’t observed behavior, and we don’t present them as such.

Your future self looks like a stranger

Part of the reason may be wiring. In a Stanford brain-scan study, students rated whether words described themselves today, themselves ten years from now, or a celebrity. A brain area that responds when people think about themselves was less active for the future self than for the present one. The bigger that drop, the more a person later took a smaller amount of money now over a larger amount later.5 It was a small study of 18 people, and the link was fragile, so treat it as a clue rather than a law.

Put that inside a company. The quarter belongs to you. The future belongs to someone your brain treats as a stranger: a successor, a later customer, a version of you who’ll have moved on. Every barrier in this report got built in that gap.

The people who could say something often don’t

The people closest to a problem usually see it first. Whether that knowledge reaches anyone with authority depends on whether speaking up is safe. In the Ethics & Compliance Initiative’s 2023 global survey, 46% of employees who reported misconduct said they experienced retaliation.33 In the 2024 federal employee survey, 72.2% agreed they could report a suspected violation without fear of reprisal, which leaves roughly one in four who didn’t agree.34 Research on corporate fraud finds that employees are among the most important sources of detection, and that the ones who come forward often pay a personal price.39

That changes how we assign responsibility. A technician who sees a falsified work order and keeps quiet has less authority, less information and fewer safe options than the executive who designed the incentive. Responsibility should rise with knowledge, authority and real alternatives. It shouldn’t vanish into “the system,” and it shouldn’t be spread evenly across everyone inside it.

The next workforce is watching

Deloitte projects that Gen Z and millennials will make up 74% of the global workforce by 2030. In its 2025 survey, 41% of Gen Z and 40% of millennials said they had already turned down a potential employer that clashed with their ethics or beliefs.6

One caution. The National Academies reviewed the research on generational labels and found they don’t support sweeping claims about what “a generation” wants.7 We don’t use these numbers to stereotype anyone. We use them for a simpler point: the people you’ll need to hire are telling you, in large numbers, that they check.

5 · The other side

The strongest case against our argument

An argument about systems that defend themselves has to be willing to be wrong. Here’s the best case against ours.

Some barriers are the part that still works

In 1960, Frances Oldham Kelsey, a new medical reviewer at the FDA, refused to approve thalidomide. The sponsor pressed her repeatedly. She held out because the safety evidence was inadequate. The drug went on to cause severe birth defects in thousands of children in countries where it was sold; it was never approved for that use in the United States. FDA approved it in 1998 for a different, tightly restricted indication.26,27

Kelsey was a barrier. She was also the institution working exactly as it should. Removing her resistance wouldn’t have been progress. So the test can’t be whether an arrangement gets in someone’s way. It has to be whether the restriction serves a defensible purpose at a proportionate cost, and whose risk it reduces.

Continuity has real value

Institutions pile up routines, assets and authority partly because coordination is hard. Splitting control can create delay, underinvestment and responsibility without capacity. Investors need returns because many investments fail. Long contracts can finance infrastructure that short-term bargaining never would. A remedy that destroys the provider can hurt the people it was meant to protect. Bell is the reminder: a structure criticized for exclusion was also one of the most productive research engines of the century.

We studied failures, so failures are what we found

Our cases were selected around scandals, audits and court records. That underrepresents the routine success nobody investigates, the preventive work that leaves no disaster behind and the reforms that happened quietly before anyone had to fight. We can’t tell you how common this pattern is across all organizations. Nobody can yet, from this kind of evidence, and we won’t pretend to.

Real reform can still fall short, and slide back

PG&E’s probation was real oversight with a real judge, and in his own words the judge had to “acknowledge failure.” Wells Fargo’s release from its asset cap was earned under a defined standard, and it still isn’t proof the culture changed for good.

The Veterans Health Administration shows both sides. After a system-wide overhaul in the 1990s, a study in the New England Journal of Medicine found the VA outperformed Medicare fee-for-service on 12 of 13 quality measures.35 In 2014, the VA Inspector General reported inappropriate scheduling practices at the Phoenix facility that understated how long veterans were waiting.36 The same organization improved, then let a measure turn into a target, then gamed it. Reform isn’t a finish line.

“Only outsiders can fix it” is wrong too

Kelsey was an insider. The credit card reforms we describe in the next section changed an industry without dismantling it. The evidence doesn’t support the idea that institutions can only be reformed from outside, or that every reform has to be paid for by someone else. It supports something narrower: a reform has to change who measures, who decides or what triggers payment, from wherever it starts.

6 · What actually changes it

Three levers, and they all move structure

The reforms that worked in our cases didn’t depend on better people. They changed at least one of three things.

Who measuresLever 1

Independent commissioning. In building commissioning, a separate party checks whether a building actually performs as designed. Lawrence Berkeley National Laboratory pooled data on 643 buildings and found more than 10,000 energy-related problems. Median energy savings were 16% in existing buildings, with a median payback of 1.1 years; new construction saved 13%, with a 4.2-year payback.29 The projects were self-selected, and later updates found smaller savings, so treat these as an upper range.

A neutral party for bad news. Since 1976, aviation’s Aviation Safety Reporting System has been run by NASA, not by the FAA that enforces the rules. Reporters get limited protection from penalties. Intake averaged about 400 reports a month in the early days. NASA now says it generally exceeds 10,000 a month.32 When the party that receives the report doesn’t control the penalty, people report.

Who decidesLever 2

The andon cord. When Toyota and GM reopened GM’s Fremont plant as NUMMI in 1984, about 85% of the workforce came from the old plant.30 Absenteeism at the old plant had run above 20%. At NUMMI it fell to roughly 2% to 4%.31 One of the biggest changes: any worker could stop the line when something was wrong. Same people, different decision rights.

The caution is part of the lesson. GM struggled to spread what it learned at NUMMI to its other plants, and NUMMI closed in 2010. A local fix doesn’t change the larger system on its own.

What triggers paymentLever 3

Binding limits beat disclosure. The 2009 CARD Act did two different things to credit cards. It limited certain fees, and it required statements to show how long minimum payments would take. Researchers estimated the fee limits saved consumers $11.9 billion a year, with no detected rise in interest charges or drop in credit to offset it. The disclosure moved the share of accounts paying enough to clear the balance in 36 months by about half a percentage point.28 The $11.9 billion is a transfer from issuers to borrowers, not new output, and it’s an estimate for the study period. But the contrast holds: changing what earns money moved behavior far more than telling people the facts.

Growth tied to the fix. Alongside penalties from other agencies, the Fed used a different lever with Wells Fargo: it capped the bank’s growth until governance changes were made and accepted. The lever was tied to growth, not just to a one-time payment.

Five stages, and where reform stalls

We judge every reform on the same five stages. Most stall before the fourth.

  1. 01AnnouncedA proposal, a memo, a press release.
  2. 02In forceLegally or contractually binding.
  3. 03ImplementedActually being done on the ground.
  4. 04MeasuredImprovement shown by someone independent.
  5. 05SustainedLasting, with an honest account of who paid.

A court order can reach stage two or three without reaching four. Calling everything short of stage five “symbolic” hides real differences. Calling stage one a success hides more.

For government contracting: six levers already in your hands

Much of our evidence comes from federal contracting, and so do many of our clients. These levers already exist in the rules. The question is whether anyone uses them.

  • Award and incentive fees tied to outcomes someone other than the contractor verifies, with no rollover of unearned fee.11,12
  • Performance evaluations (CPARS) written on time, specifically and with evidence. Contractors should want accurate ratings, even unflattering ones.
  • Independent commissioning contracted separately from the installer and fully scoped.29
  • Third-party deficiency and near-miss reporting routed to a party that doesn’t control fees or ratings. ASRS is the model.32
  • Stop-work authority that is honored in practice, not just written into the safety plan.
  • A reprisal-safety pulse. Ask your people, anonymously, whether they can report a violation without fear of reprisal. Then act on the answer.34

7 · The cost of staying

The window keeps getting shorter

Go back to the opening number. 33 years, then 21, with a forecast of 14. Whatever the exact figure turns out to be, the time an arrangement can coast on its past keeps shrinking.

Put the pieces side by side. Executives say they’ll trade future value for a smooth quarter.3 Their brains engage less with their own future selves than with who they are today.5 The workforce they’ll need is screening employers for exactly this.6 And the companies at the top of the market are being replaced faster than they used to be.2

None of those findings proves the others. Together they describe a trade that looks cheap on the business clock and expensive on the development clock. A system that protects its problem can look healthy for years. It earns, it reports and it passes its own reviews. What it stops doing is getting better at the thing it exists for, and that’s exactly the capability a shorter window punishes.

The quarter is you. The future is a stranger. Every barrier gets built in that gap.

We’ve labeled this section as interpretation on purpose. It’s where the evidence points, not something any one study proves. That’s why the next section is a set of tests rather than a verdict.

8 · The self-check

Four questions, each with a test

Asking a hard question on its own changes less than you’d hope. Research on the “question-behavior effect” finds that being asked about a behavior nudges people toward it, but the effect is small, and smaller still when the behavior means giving something up.40 So every question here comes paired with something to do.

The mirror

If nobody outside your chain checked your numbers, would they still look this good?

Test: Name the last performance number that someone outside your reporting line verified, someone who doesn’t share your bonus. If you can’t name one, commission one this quarter. Start with the number your rewards depend on most.

Where did the value go?

Test: Pick one dollar of last year’s revenue and trace it to an outcome the customer would recognize as theirs: a repair that stayed fixed, a risk that went away, a cost that came down. If the trail ends at an activity instead of an outcome, you’ve found where to look.

Are you building something that outlives you, or is it just sweet right now?

Test: Write down what would happen if the problem you’re paid to solve were solved for good tomorrow. Whose budget shrinks? Whose role ends? Whose contract goes away? If the honest answer is “ours,” decide now what you’d build next, before the market decides for you.

Are we building toward the future, or walking into our own demise?

Test: List every project, repair or investment you delayed in the last twelve months to protect a number. Put the list on the agenda of your next leadership meeting, and decide on each one in the open.

Answer them alone. Then answer them in front of your team.

Where to start, by role

Owners and founders

You set what earns money. Check whether your pricing rewards the outcome or the activity. If you’re paid by the visit, ask whether you’d still win the work if you were paid by the fix.

Operators and managers

You see the gap between the record and the work first. Keep the records accurate, raise specific discrepancies through the channels you have, and document the tradeoffs you’re forced to make.

Boards and investors

Formal independence isn’t effective oversight. Ask for raw outcome evidence, not just management’s summary, and ask how long bad news takes to reach you.

Rising leaders

You’ll inherit these arrangements. Learn which rung of the ladder each one sits on before you defend it, because the day you defend it, it becomes yours.

9 · How we did this

Eleven reports, and the limits we kept

This report and the carousel before it are the short versions of an eleven-part research series Rigby Consulting & Firm ran in 2026.

  1. Reports 1–8Eight research tracks on how systems built to solve a problem end up protecting it: history and economics, ownership and control, organizations, governance, government, technology, human behavior and consequences.
  2. Report 9A claim-by-claim verification of the first eight against primary sources. It corrected or removed claims that didn’t hold up.
  3. Report 10The synthesis: fourteen documented cases, the three conditions, the reform stages and a register of 32 publication claims, each marked ready, qualified or rejected.
  4. Report 11How leaders respond to evidence like this: what makes people defensive, and what makes them reflect.

Every claim in the series carries a label:

Fact on the recordEstimate from a research designInterpretation oursJudgment a value callOpen not settled

The limits

  • We name intent only where a court, a plea or an admission named it. A bad outcome or a benefit to someone isn’t evidence of a plan.
  • Survey figures are what people said, not what they were observed doing.
  • The cases are selected around failures and public records. They show where the pattern appears. They can’t tell you how common it is.
  • Some research was checked at the abstract level rather than reproduced from raw data. We say so where it matters.
  • Legal and regulatory statuses are current to November 2026. Matters can change after that.
  • We left out claims that sound good and don’t hold. Among them: a widely quoted 82% whistleblower-retaliation figure drawn from a narrow sample, prevalence fractions coded from our own selected cases, and several popular corporate-generosity stories that don’t survive the record.

About this report

This is analysis and opinion from Rigby Consulting & Firm, built on public records and published research. Where we describe a named company, agency or person, we rely on what the cited record says, and we label our own interpretation as ours. Pleas, settlements and agreements are described by their terms; a deferred prosecution agreement or an audit finding is not a conviction, and we don’t present one as such. Statuses are current to November 2026 and can change. Nothing here is legal, financial or investment advice.

If you find an error, tell us. A report about systems that defend their numbers should be the first to change its own.

10 · Sources

Everything cited above

  1. Innosight. Corporate Longevity: Turbulence Ahead for Large Organizations. 2016.
  2. Innosight. 2021 Corporate Longevity Forecast: Creative Destruction. 2021.
  3. Graham, J. R., Harvey, C. R. & Rajgopal, S. The economic implications of corporate financial reporting. Journal of Accounting and Economics 40, 3–73. 2005.
  4. Dichev, I., Graham, J., Harvey, C. R. & Rajgopal, S. Earnings quality: Evidence from the field. Journal of Accounting and Economics. 2013.
  5. Ersner-Hershfield, H., Wimmer, G. E. & Knutson, B. Saving for the future self: Neural measures of future self-continuity predict temporal discounting. Social Cognitive and Affective Neuroscience 4(1). 2009.
  6. Deloitte. 2025 Gen Z and Millennial Survey. 2025.
  7. National Academies of Sciences, Engineering, and Medicine. Are Generational Categories Meaningful Distinctions for Workforce Management? Report highlights. 2020.
  8. U.S. Department of Justice. Plea agreement, United States v. Balfour Beatty Communities LLC. December 2021.
  9. U.S. Senate. Senate report on privatized military housing.
  10. U.S. Government Accountability Office. GAO-23-105377, Military Housing: DOD Can Further Strengthen Oversight of Its Privatized Housing Program. April 2023.
  11. U.S. Government Accountability Office. GAO-06-66, Defense Acquisitions: DOD Has Paid Billions in Award and Incentive Fees Without Regard to Acquisition Outcomes. December 2005.
  12. U.S. Government Accountability Office. GAO-09-630, Federal Contracting: Guidance on Award Fees Has Led to Better Practices but Is Not Consistently Applied. May 2009.
  13. DoD Office of Inspector General. DODIG-2019-060, Review of Parts Purchased From TransDigm Group, Inc. February 2019.
  14. DoD Office of Inspector General. DODIG-2022-043, Audit of the Business Model for TransDigm Group Inc. and Its Impact on Department of Defense Spare Parts Pricing. December 2021.
  15. FERC, NERC and Regional Entity staff. The February 2021 Cold Weather Outages in Texas and the South Central United States. November 2021.
  16. Texas Department of State Health Services. February 2021 Winter Storm-Related Deaths, Texas. December 2021.
  17. U.S. District Court, N.D. Cal. (Judge William Alsup). Order at the end of PG&E’s probation. January 19, 2022.
  18. U.S. Attorney’s Office, S.D.N.Y. Criminal charges against General Motors and deferred prosecution agreement. September 2015.
  19. Independent Directors of the Board of Wells Fargo & Company. Sales Practices Investigation Report. April 2017.
  20. Board of Governors of the Federal Reserve System. Federal Reserve Board removes asset cap on Wells Fargo. June 3, 2025.
  21. Board of Governors of the Federal Reserve System. Termination of the 2018 enforcement action against Wells Fargo. March 5, 2026.
  22. MACPAC. State-Reported Medicaid Unwinding Data. November 2024.
  23. Gupta, A., Howell, S. T., Yannelis, C. & Gupta, A. Does Private Equity Investment in Healthcare Benefit Patients? Evidence from Nursing Homes. NBER Working Paper 28474, revised.
  24. Cleary, E. G., Beierlein, J. M., Khanuja, N. S., McNamee, L. M. & Ledley, F. D. Contribution of NIH funding to new drug approvals 2010–2016. PNAS. 2018.
  25. Watzinger, M., Fackler, T. A., Nagler, M. & Schnitzer, M. How Antitrust Enforcement Can Spur Innovation: Bell Labs and the 1956 Consent Decree. American Economic Journal: Economic Policy. 2020.
  26. U.S. Food and Drug Administration. Frances Oldham Kelsey: Medical reviewer famous for averting a public health tragedy.
  27. U.S. Food and Drug Administration. Thalidomide approval letter, NDA 20-785. July 16, 1998.
  28. Agarwal, S., Chomsisengphet, S., Mahoney, N. & Stroebel, J. Regulating Consumer Financial Products: Evidence from Credit Cards. Quarterly Journal of Economics 130(1). 2015.
  29. Mills, E. Building Commissioning: A Golden Opportunity for Reducing Energy Costs and Greenhouse Gas Emissions. Lawrence Berkeley National Laboratory. 2009.
  30. Helper, S. & Henderson, R. Management Practices, Relational Contracts, and the Decline of General Motors. Journal of Economic Perspectives 28(1). 2014. Also HBS Working Paper 14-062.
  31. Adler, P. S. Time-and-Motion Regained. Harvard Business Review. January 1993.
  32. NASA Aviation Safety Reporting System. ASRS Program Briefing and CALLBACK issue 555: 50 Years of Turning Safety Reports Into Safer Skies. April 2026.
  33. Ethics & Compliance Initiative. Global Business Ethics Survey 2023. 2023.
  34. U.S. Office of Personnel Management. Federal Employee Viewpoint Survey, 2024 results. 2024.
  35. Jha, A. K., Perlin, J. B., Kizer, K. W. & Dudley, R. A. Effect of the Transformation of the Veterans Affairs Health Care System on the Quality of Care. New England Journal of Medicine 348. 2003.
  36. VA Office of Inspector General. Review of Alleged Patient Deaths, Patient Wait Times, and Scheduling Practices at the Phoenix VA Health Care System. August 2014.
  37. Liebman, J. B. & Mahoney, N. Do Expiring Budgets Lead to Wasteful Year-End Spending? Evidence from Federal Procurement. American Economic Review 107(11). 2017.
  38. DeCelles, K. A., DeRue, D. S., Margolis, J. D. & Ceranic, T. L. Does power corrupt or enable? When and why power facilitates self-interested behavior. Journal of Applied Psychology 97(3). 2012.
  39. Dyck, A., Morse, A. & Zingales, L. Who Blows the Whistle on Corporate Fraud? Working paper; published in Journal of Finance, 2010.
  40. Wood, C., Conner, M., Miles, E., Sandberg, T., Taylor, N., Godin, G. & Sheeran, P. The impact of asking intention or self-prediction questions on subsequent behavior: A meta-analysis. Personality and Social Psychology Review 20(3). 2016.
  41. Allcott, H. Social norms and energy conservation. Journal of Public Economics 95. 2011.
  42. U.S. Government Accountability Office. Bid Protest Annual Report to Congress for Fiscal Year 2025.
  43. Federal Acquisition Regulation, Subpart 42.15, Contractor Performance Information.
  44. Acquisition.gov. FAR Overhaul: Frequently Asked Questions.
  45. Bank of England. The collection of slavery compensation, 1835–43. Staff Working Paper. 2022.
  46. Delaware Supreme Court. In re Tesla, Inc. Derivative Litigation. December 19, 2025.
  47. Microsoft. The next phase of the Microsoft–OpenAI partnership. April 27, 2026.

Sources 9, 38 and 41 to 47 informed the series and the synthesis behind this report (how the deck speaks to people with authority, peer-norm effects, protest statistics, contract rules and the remaining documented cases) without being quoted above. Statuses current to November 2026.

We aren’t invested in the old numbers.

Rigby Consulting & Firm is an outside set of eyes. We help you see which clock your system is running on, and what it would take to change it. The decision stays yours.

Start a conversation

© 2026 Rigby Consulting & Firm Limited Liability Co. · rigbyfirm.com