You will eventually face a decision where the legal answer and the right answer aren't the same, a layoff timed to flatter a quarter, a defect you could quietly ship, a customer's data you could use in a way they never imagined. "Business ethics" is the study of how to handle exactly those moments. Ethical frameworks are the small set of reusable tests that turn a queasy gut feeling into a defensible decision you can explain out loud.

The quick version

  • Business ethics is applied ethics: working out what a company should do, not just what the law lets it do. The law is the floor, not the ceiling.
  • There are three classic frameworks for testing a decision, by its consequences (who is helped or harmed), by its duties and rights (what we owe regardless of outcome), and by character (what a person of integrity would do). Run a hard call through all three.
  • The big strategic question is who the business is for, only shareholders (Friedman) or all stakeholders including staff, customers and communities (Freeman). Most modern boards land somewhere between.
  • Ethics is a system, not a slogan. A values poster changes nothing; incentives, escalation routes and what leaders tolerate change everything.

The idea in depth: three lenses, one decision

Philosophers have argued about right and wrong for millennia, but for a working leader the inheritance boils down to three practical lenses, each rooted in a distinct tradition in moral philosophy. They line up well with the approaches the Markkula Center for Applied Ethics at Santa Clara University teaches in its widely used Framework for Ethical Decision Making, Markkula actually breaks the ground into more lenses than three (utilitarian, rights, justice, common good, virtue and care), but for a decision you have to make this afternoon, grouping them into consequences, duties and character is the version that sticks.

The first is consequentialism (its best-known form is utilitarianism): the right action is the one that produces the greatest balance of good over harm across everyone affected. The second is deontology, the ethics of duty and rights: some things, honesty, keeping promises, not treating people merely as means, are obligations regardless of the payoff. The third is virtue ethics, which asks not "what should I do?" but "what would a person of good character do here, and what does this choice make me into?"

None of the three is sufficient alone, which is the point. Consequences-only thinking can justify almost anything if the spreadsheet looks good enough. Duty-only thinking can be rigid to the point of cruelty. Character-only thinking is vague at 4pm on a Thursday when you need an answer. So run a genuinely hard decision through all three in turn. Who is helped and harmed, and how badly? Whose rights or reasonable expectations would this breach? Would I be comfortable if this appeared, in full, on the front page with my name on it? When the three lenses agree, act with confidence. When they conflict, you have found the actual ethical problem, and naming it is most of the work.

flowchart TD
  A(["A hard decision
(legal, but uneasy)"]) --> B(["Consequences lens
who is helped / harmed?"])
  A --> C(["Duties & rights lens
what do we owe, regardless?"])
  A --> D(["Character lens
front-page test"])
  B --> E{"Do the three
lenses agree?"}
  C --> E
  D --> E
  E -->|"Yes"| F(["Act with confidence"])
  E -->|"No"| G(["You've found the real
ethical problem, name it,
escalate it, don't bury it"])
The three-lens test, adapted from the Markkula Center framework, agreement is permission to act; conflict is the signal to slow down. Leaders Loop

An honest limitation. These frameworks clarify a decision; they do not make it for you. Two thoughtful people can run the same call through all three lenses and still disagree, because they weigh harms differently or rank duties in a different order. The frameworks are not a calculator that outputs "ethical: true." Their real value is forcing the considerations into the open so the decision is reasoned and reviewable rather than rationalised after the fact, which is exactly what an audit, a regulator, or your own future self will ask for.

Who is the business actually for?

Underneath day-to-day dilemmas sits one strategic question that shapes all of them: to whom does a company owe its primary duty? The two landmark answers define the debate.

In 1970, economist Milton Friedman argued in The New York Times Magazine that "the social responsibility of business is to increase its profits", that an executive is an agent of the shareholders who own the firm, and that spending their money on social causes is, in effect, taxing them without consent. Crucially, even Friedman bounded this within "the rules of the game", open, fair competition "without deception or fraud." His doctrine isn't a licence to cheat; it's an argument about whose interests a manager is hired to serve.

Fourteen years later, R. Edward Freeman offered the counter-argument in Strategic Management: A Stakeholder Approach (1984). A company, Freeman held, creates value not only for shareholders but for employees, customers, suppliers and the communities it touches, and treating those groups well isn't charity siphoned from profit, it's how durable value gets built in the first place. Stakeholder theory became the academic backbone of modern corporate responsibility.

The law tells you what you can be punished for. Ethics tells you what you'll have to defend.

This isn't a settled debate, but the centre of gravity has moved. In August 2019 the US Business Roundtable, an association of major-company chief executives, issued a statement signed by 181 CEOs redefining the purpose of a corporation around delivering value to all stakeholders, a deliberate break from the shareholder-primacy language it had used since 1997. The leadership task is to make your own answer explicit before a crisis forces one on you. Decide, in writing, which trade-offs you will and won't make, and where genuine conflict between groups should surface to the board rather than be quietly resolved in profit's favour by whoever happens to be in the room. A note of honesty: critics argue the Roundtable statement changed rhetoric more than behaviour. Declaring a stakeholder purpose is the easy part; wiring it into incentives is the hard part, which is the next idea.

Ethics is a system, not a poster

The most useful thing to internalise is that organisational ethics is mostly an engineering problem, not a virtue problem. Good people make bad choices inside systems that quietly reward the bad choice, a sales commission that pays out before the customer is actually served, a target that can only be hit by cutting a corner, a culture where raising a concern marks you as "not a team player." If you want ethical behaviour, the durable lever is not exhortation; it is the incentives and the escape routes you design.

This is why psychological safety belongs in any serious conversation about ethics: people only flag the defect, the dodgy deal or the harassment if they believe speaking up won't end their career. The practical apparatus follows, a code of conduct that names hard cases (not just "act with integrity"), a channel that is genuinely safe to use, named owners for escalation, and the habit of running consequential decisions through the three lenses before they're made. So audit your incentives for the places where doing the right thing costs an employee personally, then remove that cost. That gap is where your next ethical failure is already waiting.

A worked example

Take a fictional logistics platform, call it Cartwheel. (Illustrative throughout; not a real company.) An engineer discovers that a pricing algorithm, entirely legally, charges more to customers in postcodes with fewer competing options, disproportionately lower-income areas. It lifts margin by an estimated 3% (an illustrative figure). Legal signs off: no rule is broken. The growth lead wants to ship.

Run it through the three lenses. Consequences: a modest margin gain for the company against a real harm concentrated on customers least able to absorb it, plus serious downside if it surfaces publicly. Duties and rights: it treats vulnerable customers as a means to a margin target and trades on their lack of alternatives, a breach of the fair-dealing most people would say a customer is owed. Character: the front-page test fails instantly; no one wants to defend "we charged poor people more because they couldn't shop around."

flowchart LR
  A(["Legal sign-off:
nothing is broken"]) --> B(["Consequences:
+3% margin vs
concentrated harm"])
  A --> C(["Duties & rights:
exploits lack of choice"])
  A --> D(["Character:
fails front-page test"])
  B --> E(["All three lenses
point the same way"])
  C --> E
  D --> E
  E --> F(["Don't ship it,
and ask why the incentive
rewarded shipping it"])
"Legal" cleared it; the three lenses didn't. The framework gives the team language to say no, and a reason a board will back. Leaders Loop

All three lenses converge: don't ship it. But the deeper lesson is systemic. The engineer who raised it could only do so because Cartwheel made it safe to. And the growth lead wanted to ship not from malice but because the dashboard rewarded margin and was silent on fairness. The lasting fix isn't a stern email, it's changing what the incentive measures, so the next algorithm is built with the constraint baked in rather than caught at the last moment by one brave person.

Frequently asked questions

Isn't "business ethics" just obeying the law?

No, the law is the floor. Plenty of profitable behaviour is perfectly legal and clearly wrong, and plenty of ethical questions (how you weigh staff against shareholders, what you do with data you're permitted to use) the law simply doesn't address. Treat compliance as the minimum bar to clear, and ethics as the standard you'll actually be judged by, by customers, employees and your own conscience.

Which ethical framework is the "right" one?

None of them, used alone. Consequences, duties, and character each catch something the others miss, which is why the practical approach is to run a hard decision through all three rather than pick a favourite. When they agree, you can act with confidence; when they conflict, you've located the genuine dilemma, and that's the moment to slow down, write down the trade-off, and escalate it rather than resolve it quietly.

Do ethics and profit actually trade off?

Sometimes, honestly, yes, anyone claiming they never conflict is selling something. But the framing of a permanent trade-off is often wrong over the long run: trust, reputation and the ability to recruit and retain good people are economic assets, and they're built by exactly the behaviour ethics asks for. Stakeholder theory's core claim is that treating people well is largely how durable value is created, not a tax on it. Where a real short-term conflict exists, name it openly rather than pretending it away.

How do I stop our code of conduct being a poster nobody reads?

Make it about hard cases, not platitudes. "Act with integrity" guides no one; "here's what to do when a customer offers you a gift / when a target can only be hit by cutting a corner / when you suspect a colleague is fiddling figures" does. Then pair it with a channel that's genuinely safe to use and leaders who visibly back the person who raises a concern. Behaviour follows what's rewarded and tolerated far more than what's framed on the wall.

What's the single highest-leverage thing a leader can do?

Audit your incentives for places where doing the right thing costs an individual personally, a commission paid before the customer is served, a metric that ignores how a number was hit, a culture where whistleblowers get sidelined. Ethical failures cluster exactly there. Removing that personal cost does more for everyday ethics than any training module.

Related in the Toolkit

Ethics doesn't sit on its own, it's the layer beneath how a company defines responsibility (corporate social responsibility) and how a board actually governs the trade-offs (board roles & committees).

Where to go next