---
title: "Misleading advertising as a market failure (Heath, VW)"
description: "Why misleading advertising undermines what makes profit legitimate, per Heath: the Volkswagen case from US DOJ, a four-question test and its limits."
url: "https://staging.khoi.co.uk/en/blog/quang-cao-hieu-lam-hong-thi-truong"
lang: en-GB
author: "Lê Quốc Khôi"
date: 2026-11-03
updated: 2026-11-03
topics: ["Ethics & corporate governance"]
tags: ["Market failure", "Information asymmetry", "Externality", "Misleading advertising", "Business ethics"]
---

# Misleading advertising breaks markets, not only ethics

## Key takeaways

- Heath (2006): profit is legitimate because competition pushes markets towards Pareto efficiency, so the ethical firm does not seek to profit from market failure, even where the law does not forbid it (Heath, 2006, p. 550).
- A US Department of Justice release of 11 January 2017 says Volkswagen AG agreed to plead guilty to three felony counts and pay a US$2.8bn criminal penalty and US$1.5bn in civil settlements, US$4.3bn in all, for selling about 590,000 "clean diesel" cars with software that cheated emissions tests (US Department of Justice, 2017).
- The same conduct touches three conditions: information, externalities (NOx up to 40 times the US standard) and competition (not in Heath's list) (US Department of Justice, 2017). The classification is mine.
- The four-question test for a marketing claim is my construction from these conditions. Limits: law is a blunt instrument (Heath, 2006, p. 550), and fixing one condition at a time need not improve the whole (Lipsey and Lancaster, 1956, p. 12).

On 11 January 2017 the US Department of Justice announced that Volkswagen AG had agreed to plead guilty to three felony counts and pay US$4.3bn, for selling roughly 590,000 diesel cars marketed as "clean diesel" with software that cheated emissions tests (US Department of Justice, 2017). Whoever signs off marketing copy at a far smaller company meets a miniature of the same question: how true is the line about to be printed on the box, and when the sales director asks "where is the law broken?", whom does "but that is unethical" persuade?

## Why is "it is unethical" a weak argument in a company meeting?

Because in the meeting it sounds like a cost to be weighed, not a reason to stop. Heath cites the view of morality as a "principled constraint on the pursuit of self-interest", which turns business ethics into a constraint on profit rather than an order to maximise it (Heath, 2006, p. 541). The reply "doing right pays in the end" does not hold either: he notes that in certain cases cheating customers can be profitable (Heath, 2006, p. 542). In my inference, the debate then reduces to who prices the ethical cost, and whoever holds the revenue chart usually wins.

## What is Heath's alternative argument?

Heath defends profit on Paretian grounds: profit is valued because it generates the competition that pushes prices towards market-clearing levels (Heath, 2006, p. 541). The ethical firm therefore does not profit from places where the price system is broken.

[Pareto efficiency](https://staging.khoi.co.uk/en/glossary#pareto-efficiency) means no one can be made better off without making someone else worse off. Competition delivers it only if there are no externalities, buyers and sellers hold symmetric information, insurance markets are complete, and agents are rational with consistent preferences (Heath, 2006, pp. 549-550). Those conditions are never fully met, so firms can profit from "non-preferred competitive strategies" (rather than price, quality or innovation) such as producing pollution or selling products with hidden quality defects (Heath, 2006, pp. 549-550). That is [market failure](https://staging.khoi.co.uk/en/glossary#market-failure).

Outside the firm, Heath's specific ethical duty is to refrain from such strategies even when the law allows them: "the ethical firm does not seek to profit from market failure" (Heath, 2006, p. 550). The law can prohibit only flagrant misinformation, so "misleading advertising stands to false advertising as deception does to fraud" (Heath, 2006, p. 551). Profiting from it is objectionable not because it breaches a duty of loyalty to the customer as a stakeholder, but because "it undermines the social benefits that justify the profit orientation in the first place" (Heath, 2006, p. 551).

That argument speaks the language of the meeting. It does not say "do not make money"; it names the way of making money that destroys the reason you are allowed to. The library calls this [Market Failures Approach (MFA)](https://staging.khoi.co.uk/en/framework/market-failures-approach).

## What does the US Department of Justice say about Volkswagen?

The charges against six individuals and the civil claims in the 11 January 2017 release are allegations only, except where VW admitted them (US Department of Justice, 2017).

The US$2.8bn criminal penalty relates to a long-running scheme to sell about 590,000 diesel vehicles in the US using a "defeat device" to cheat the EPA and CARB tests. The civil resolutions added US$1.5bn, so 2.8 + 1.5 = US$4.3bn. At that date the plea agreement still had to be accepted by the court (US Department of Justice, 2017).

When the software detected a test, the car met the US NOx standard; otherwise its emissions controls were substantially reduced and NOx reached up to 40 times the standard. From model year 2009 to 2016 the cars were marketed as "clean diesel" and environmentally friendly, while the software was hidden from the EPA, CARB and customers. According to the charging documents, once the alleged conspirators realised they could not design a diesel engine that both met the stricter NOx standards and attracted enough US demand, they decided to use software to cheat the tests (US Department of Justice, 2017).

| Condition | What the release says | How I classify it |
|:--|:--|:--|
| Symmetric information | Software hidden from customers; cars sold as "clean diesel" | Buyers could not check what they were promised. |
| No externalities | NOx up to 40 times the standard; US Attorney McQuade said the conduct "harmed our clean air" (US Department of Justice, 2017) | The cost fell on people breathing the air. |
| Fair competition (framework) | Also McQuade: it "created a competitive disadvantage for other automakers who play by the rules" (US Department of Justice, 2017) | A non-preferred strategy beat rivals who followed the rules. |

This is not a hard case. DOJ treats it as fraud, and Heath says that in many cases profiting from market failure is illegal precisely because the state has tried to ban such strategies (Heath, 2006, p. 550). The line of copy you review will usually sit in the zone that is not banned but not preferred either. See also the library's [Volkswagen case](https://staging.khoi.co.uk/en/case/volkswagen-emissions-scandal). The release does not say whether VW made a net profit or loss.

## How do you review one of your own claims with this framework?

Ask four questions, each tied to a condition. The set is my construction, not a list Heath gives.

1. Can the buyer check this before or soon after buying? If they can only trust you, information sits on one side.
2. How will an ordinary reader understand the line, and is that true to what you know? Literally true but misleading is the zone the law struggles to reach.
3. Does any cost of this claim, or of what it hides, fall on someone outside the transaction?
4. Would a rival telling the plain truth lose out because of this line?

A "yes" to question 1 or 2 together with a "yes" to question 4 is the strongest warning sign.

**Invented example.** Firm M sells a household air purifier; the firm, the figure and the test conditions are all my inventions. The box says "Removes 99.9% of fine dust", measured in the firm's small sealed test chamber and not in a real living room. Q1: nobody at home has a meter to compare with the box. Q2: readers take it to mean "in my room, the purifier removes nearly all the dust", which may not be true. Q3: the cost falls mainly on the buyers themselves, so the externality is small. Q4: yes, because a rival printing "measured in a 20 square metre room" with a lower figure would look worse despite a good machine. Fix: print the figure together with the test conditions.

## What would the four ethical lenses say about the same claim?

Each lens gives an answer, but none supplies a ready-made business reason. The library's [Four ethical lenses](https://staging.khoi.co.uk/en/framework/four-ethical-lenses) page poses four questions; the application to Firm M's box is mine.

- Rights (am I using anyone merely as a means?): an asthmatic buyer is used as a means to a sale if the decision rests on a gap they cannot close.
- Consequences (what maximises total benefit?): revenue and jobs against wasted purchases and health.
- Contract (which rule could everyone accept?): buyers would struggle to accept "sellers may print laboratory figures without conditions", since once everyone does it they discount every number. This lens sits closest to Heath.
- Virtue (what would a decent person do?): an honest seller would not print it, but the lens is vague at the margins.

Heath differs by deriving constraints from the policy aims behind the regulatory environment in which firms compete, not from general morality (Heath, 2006, p. 551). The framework page advises picking one main lens and using another to challenge it.

## How far does this approach go, and what does it leave open?

It gives a business reason to stop some claims and leaves three things open.

First, Heath himself calls the law a blunt instrument, because the state lacks information and enforcement is costly (Heath, 2006, p. 550). The boundary between misleading and mere exaggeration still needs judgement. The 2006 paper does not discuss purely emotional brand advertising, so the four questions cannot settle that kind.

Second, the approach does not impose direct duties to every stakeholder: Heath treats misleading advertising as undermining the market, not as betrayal of the customer (Heath, 2006, p. 551). My inference: an ad that is harmless to market efficiency but demeaning to viewers falls outside this framework.

Third, second best. Lipsey and Lancaster show that when one Paretian condition cannot be met, the other conditions, although still attainable, are in general no longer desirable (Lipsey and Lancaster, 1956, p. 11). A situation in which more but not all conditions are met is not necessarily better than one in which fewer are (Lipsey and Lancaster, 1956, p. 12). My inference: for a firm, that is a reason not to promise that the market will reward the one company that drops equivocal advertising. The theorem concerns welfare in general; applying it to one advertising claim is my extension.

## Going deeper: what the three sources are, what I read, and what is missing

Heath (2006), *Business Ethics Quarterly* 16(4), pp. 533-557: I read the full text in a copy hosted on silo.tips, with journal pagination, and have not compared it with the publisher's version. I could not read any page of Heath (2014), the framework's origin according to the library, so I do not use it; the 2006 paper does not discuss second best.

US Department of Justice press release, 11 January 2017: I read it in full through the Internet Archive snapshot of 10 February 2018, because the live page blocks automated access. It is a press release, not a judgment; I did not read the plea agreement or the indictment, and I did not check later developments.

Lipsey and Lancaster (1956): I read the opening, sections I, II and IV (pp. 11-13 and 17) in a scan; the formal proof I only skimmed and do not use. The four lenses come from the framework page, and I have not read Rachels and Rachels (2012).

Two inconsistencies in the library: the framework page's list of conditions differs from the 2006 paper, and the Volkswagen case is titled 2015-2017 while the release describes a scheme running from 2006 to model year 2016.

## Frequently asked questions

### Is a claim that is literally true but leaves readers with the wrong impression treated like fraud?

Heath's comparison is that misleading advertising stands to false advertising as deception does to fraud, and that it would be illegal were it not for practical limits on the law (Heath, 2006, p. 551). That is a moral judgement, not a legal finding. This post does not examine the law in Vietnam or the UK, so the legal question belongs with someone qualified to answer it.

### Does a firm owe customers the same duty it owes shareholders, according to Heath?

No. Inside the organisation Heath stays close to the shareholder-focused view; towards outsiders the duty is to refrain from profiting from market failure (Heath, 2006, pp. 549-550). Citing Goodpaster, he notes that obligations to customers need not be fiduciary: a manager might have to respect certain customer rights without a fiduciary duty to advance their interests (Heath, 2006, p. 543).

### Is US$4.3bn the whole cost to Volkswagen?

No. The 11 January 2017 release says the civil settlement with the EPA sat on top of a US$14.7bn settlement announced in June 2016 for the 2.0-litre cars and a US$1bn settlement announced in December 2016 for the 3.0-litre cars, which together included nearly US$3bn for environmental mitigation (US Department of Justice, 2017). If all US$4.3bn is additional, the three figures sum to US$20.0bn (my addition), but the release does not split what went to owners and what went to the environment.

### Does the second-best theorem mean one should not fix a single market failure?

No. It says that when one Paretian condition cannot be met, satisfying the remaining conditions is in general no longer desirable (Lipsey and Lancaster, 1956, p. 11). For someone reviewing advertising, my inference is that the sensible consequence is not to promise colleagues that the market will certainly reward honesty.

## References

- Heath, J. (2006) ‘Business ethics without stakeholders’, *Business Ethics Quarterly*, 16(4), pp. 533–557. https://doi.org/10.5840/beq200616448
- Lipsey, R.G. and Lancaster, K. (1956) ‘The general theory of second best’, *The Review of Economic Studies*, 24(1), pp. 11–32. https://doi.org/10.2307/2296233
- US Department of Justice (2017) *Volkswagen AG agrees to plead guilty and pay $4.3 billion in criminal and civil penalties (press release, 11 January)*. Available at: https://www.justice.gov/usao-edmi/pr/volkswagen-ag-agrees-plead-guilty-and-pay-43-billion-criminal-and-civil-penalties-and.

---

Canonical HTML version: https://staging.khoi.co.uk/en/blog/quang-cao-hieu-lam-hong-thi-truong