The billable hour is out of time

On 13 May 2007 Charlie Munger stood in front of the graduating class of the University of Southern California law school. He was a lawyer himself, and incentives were his life’s subject. What he gave the young lawyers was not encouragement.

Read closely, Munger is aiming not at billing by the hour but at the hours target. He means the incentive system, not the price. The flaw sits a level below that, though, and the target is only its most visible consequence. The hour was never meant to be a price. It was a tool to make lawyers faster. Raised to a price, it does the opposite.

The idea of timekeeping came out of legal aid

In 1913 Reginald Heber Smith was counsel to the Boston Legal Aid Society, with roughly 2,000 cases a year and almost no money. Together with a professor at Harvard Business School he built a system of accounts and case statistics. It worked. Cases closed rose by 65 percent, and the cost per case fell from $3.93 to $1.63.¹ Measurement forced efficiency, which is what it was built for.

Then it became a price. Smith, by then managing partner of a Boston firm, introduced timesheets there around 1920, the smallest unit a tenth of an hour. By 1940 he had turned this internal measuring tool into the client’s invoice. According to his firm’s history, he considered measured time “a fair, logical, transparent and indisputable method for valuing legal services”.²

That idea is the founding error. You cannot value a service by measuring the effort of the person providing it. The method was wrong not because of later excess but from the first day.

The party that made timekeeping universal was not a client but a trade association. In 1958 the American Bar Association published a pamphlet titled “The 1958 Lawyer and His 1938 Dollar”. The concern behind it was that lawyers’ incomes had risen more slowly than those of doctors and, worse, of dentists. Its recommendation was to record every working hour and to align fees with an hourly rate, on an assumption of 1,300 fee-earning hours per year.³ Later the hour had a substantive occasion too. In 1975 the Supreme Court struck down minimum fee schedules for lawyers as price fixing in Goldfarb v. Virginia State Bar.⁴ The hour took their place. But an occasion is not an argument.

The conflict is in the model, not in the people

In every other purchase both sides want the same thing. The customer wants it done quickly, and so does the supplier, because quicker means more orders. The hour reverses that. Every hour the lawyer saves costs the lawyer money.

The rest follows from there. No fee agreement, however carefully drafted, changes that.

Experience is punished. Whoever knows the answer after 20 years in 10 minutes earns less on that question than the one who needs 10 hours for it. What is paid for is not the knowledge but its absence, and the client funds the learning curve.

The pragmatic solution does not pay. The short clause instead of the long one, the call instead of the memo, the third negotiation round nobody needs: every time the better outcome for the client, every time the worse one for the invoice.

Least of all does investment in efficiency pay. Whoever maintains templates, builds clause libraries, and organizes his knowledge so he can find it again is cutting his own revenue. A model that punishes the building of tools does not get tools.

None of this is a charge against lawyers. The incentive system leads decent people into conduct they did not choose. That is exactly what Munger warns against in the quotation at the start of this article.

The target Munger warned about

Turned inward, the same incentive works more sharply still. What reaches the client as an invoice is, for the lawyer, an annual number to hit.

In August 2025 an international law firm told its associates they were expected to contribute a minimum of 2,400 “productive” hours a year, generally including at least 2,000 billable ones; the rest was to go to business development, professional development, recruiting, and firm initiatives. Whoever misses the number is no longer in good standing, and good standing decides the bonus. The formal bonus threshold of 1,950 billable hours stayed in place alongside it.⁵ Spread over 52 weeks, 2,400 hours is 46 hours in every week of the year, with no holidays and no days off.

In 2007 Munger named that exact number as the thing he could not have lived with. Eighteen years later it is the condition for being in good standing.

The reason for targets like these is mundane, and it is also the heart of the matter. Hours can be counted. What makes legal work valuable cannot: the proceedings that never happen, the negotiation that does not escalate, the call that disposes of a problem before it becomes one. Whoever needs a number takes the only quantity that has one and turns it into the target. That measures effort and calls it performance. In this business only trust counts. And trust appears in no target.

What the law demands of your lawyer, and what you get to see

The international debate on the billable hour misses a point that German courts made long ago. The time you are billed for is hard for you to check.

When your lawyer bills by time, you pay the hours he proves. They must stand in reasonable proportion to the scope and difficulty of the matter. The Federal Court of Justice accepts differences in the way lawyers work.

The way lawyers work differs from one lawyer to the next, as every client knows. Differences in the time taken are therefore in principle to be accepted, the court held.

The Federal Constitutional Court has found that the time actually spent remains hidden from the client. This lets a dishonest lawyer bill his client to a large extent without any check on his actual performance.⁶ Your lawyer therefore bills the time he needs, and you cannot see it. That comes not from consultants’ literature but from Germany’s highest court, even though the court set aside the reduction of the fee in that case. In my view, a model whose central quantity only the party billing it can observe rests on trust alone.

A time-based fee clause that contains nothing beyond an hourly rate is not plain and intelligible to consumers, the Court of Justice of the European Union held. That alone does not make it invalid.⁷ At home the standard-form 15-minute increment fell in 2020 before the Federal Court of Justice, at least against consumers,⁸ six years later before the Düsseldorf Court of Appeal against business clients as well, with the consequence that the work must be billed to the minute.⁹ At the highest level the point is still open for business clients.

Weeks afterwards the clause fell under which your silence in response to a timesheet counts as acknowledgment of the hours, expressly in dealings with business clients as well.¹⁰ The sums at stake show in the same case. The appeal court had ordered repayment of 77,905.49 euros to the client; the Federal Court of Justice set that aside and remanded. In this case the dispute was about the fee agreement, not about the hours.

The instruments that make hourly billing workable are falling one by one.

The same fight, 35 million dollars, in London

On 24 March 2026 the Senior Courts Costs Office in London opened invoices from Wilmer Cutler Pickering Hale and Dorr for 35,343,213.96 dollars to detailed assessment.¹¹ The firm is the successor to the Boston partnership whose managing partner, Reginald Heber Smith, turned the hour into a price.

Not a word of it turned on the hours. The costs judge held the invoices to be as detailed as anyone could reasonably expect of a law firm’s computerized record. He called the objections to them a quibble. What decided the case was a standard line in the invoice footer: that the invoice included only services and disbursements posted to date. On that wording no invoice was final. None was a statutory bill, the formal bill under the Solicitors Act 1974. Only the delivery of such a bill starts the time limits for assessment. For the individual invoices, those time limits therefore never began to run. The whole series of invoices became a single bill, assessable in full.

The second point goes to the hourly rate itself. The firm had raised its rates twice during the retainer without telling the client. A unilateral right to raise them by any amount and at any time, the judgment holds, does not fit with a contentious business agreement. Under such an agreement the client can no longer challenge the rates on assessment.

English law, an English costs procedure, of no direct application to a German engagement. In the two German cases of 2026 it was clauses that fell, not the fees: before the Düsseldorf Court of Appeal the increment clause, before the Federal Court of Justice the acknowledgment clause. Both times the fee agreement otherwise remained valid. In London a line in the invoice footer decided the matter.

The hourly rate is the wrong number

The most common objection is that the hour is at least transparent and comparable. Neither is true.

It would be comparable only between people of equal qualification, and nobody compares those. More importantly, the rate is only one of two factors. The other is the number of hours. You cannot see how many your lawyer actually spends. A lawyer with the lower rate can cost you twice as much. He must set out the hours in a concrete and verifiable way and prove them in a dispute.¹² But you can hardly judge how much work was needed. Whoever negotiates the rate negotiates the factor they can control and leaves the other one open.

Clients tell me regularly what this does. An acquaintance had negotiated a pointedly low hourly rate with his commercial firm. What followed was a quarter of an hour billed for every mere forwarding of an email, performed by the assistant, not the lawyer. The total reached five figures, and the engagement ended. Smith thought a tenth of an hour was the smallest sensible unit.

The incentive works in the other direction too. Whoever does not want to raise the hourly rate can just as well suggest writing down more hours. In plain terms, that is an invitation to bill time that never accrued. That the suggestion is even conceivable within the hourly model says more about the model than any study.

Finally, a point that is rarely said out loud. Which client enjoys picking up the phone when billing starts at the first second? The short call that would have prevented a problem never happens. And what high billing targets do has been measured since 2015 by the psychological study What Makes Lawyers Happy?: the higher the target, the lower the intrinsic motivation and the satisfaction, and the more frequent the alcohol abuse.¹³ Nobody wants an exhausted lawyer on their matter.

Diagram: the hourly rate, which is negotiated and visible, multiplied by the number of hours, which is not negotiated and hard to check, equals your invoice.

Hourly ratenegotiated, visible

×

Number of hoursnot negotiated, hard to check

=

Your invoice

What a US corporation refused to pay for

Some years ago a US corporation wanted to engage me. Invalid clauses across hundreds of live B2B contracts in Germany, a problem its own legal department had failed to solve. What was needed was someone to develop a strategy in the first place – legal, communicative, commercial.

The inquiry arrived together with the corporation’s fee policy, and it is worth reading closely.

It expected, first of all, a 10 percent discount off standard rates, as a rule rather than as the outcome of a negotiation. A price that is required by the fee policy to sit 10 percent below the one quoted is not a price. Either the rate was too high before, or it will be next time. Both sides know it, neither says it.

It required tenths of an hour, that is a six-minute increment. Market standard, and still the wrong unit. The solution to a contract problem does not arrive in six-minute blocks. It arrives while reading something else, in conversation, on the way to a meeting. What does divide cleanly into six-minute blocks is execution.

And it excluded payment for documents and research produced for another client and then used on this engagement. How that is to be established is not stated. More important, this is precisely the knowledge that makes a specialist faster and better. In a knowledge economy it is the real asset. The fee policy pays the lawyer who starts from zero, and it pays him for the hours.

Then the list of what would not be paid for either: preparing cost estimates and budgets, discussing staffing, conferences, instruction, training. What it would pay for was execution.

Except that the engagement consisted of nothing but what would not be paid for. Working it out was the job. A fee policy that declares estimating, discussing, and thinking together non-billable assumes that the lawyer executes what the company has already worked out. Here the opposite was true, or the inquiry would never have been sent. I declined.

Both sides are punishing the same thing. The model pays the lawyer nothing for building knowledge, and the fee policy pays him nothing for bringing it along. What is left is the hour, starting from zero.

Why the question is breaking open now

Artificial intelligence is making visible a calculation that was always wrong. Jordan Furlong put it precisely in the American Bar Association’s trade press in June 2026. Businesses in other industries welcome efficiency gains because they cut production costs.

So every client is now asking why they pay by the hour when the work takes seconds. And every lawyer is asking why they should become more efficient. Both questions have the same answer, and it is the fixed fee.

Little has moved so far. Across the legal-department budgets running through one of the large e-billing platforms, 90 percent of legal fees still flow through hourly arrangements.¹⁴

How hard the profession finds this shows in a scene two authors reported in the ABA trade press in July 2026. Asked by an in-house counsel why the firm’s much-publicized AI initiatives were not translating into lower fees, her partner answered that the technology gave him more time to think about her matters. Clients buy judgment and solutions, not hours. Why that judgment also needs to be organized inside the company is the subject of the article “AI gives answers. Your company needs the right ones”.

The strongest objection, refining the hour instead of abolishing it

The best counterposition also comes from the ABA trade press and deserves an honest answer. It holds that the hour should not be abolished but refined, with technology raising the share of valuable work inside each hour.

The example the author uses to illustrate it refutes him. A partner quotes $350,000 for a deal and then goes through the timesheet striking whatever strikes him as not valuable. That is a disguised fixed fee. The price was set in advance, on value, and the timesheet is made to fit afterwards.

What to do instead

The German statutory fee regime shows the principle, whether or not you approve of it. It ties the fee to the value of the matter (Gegenstandswert), not to the effort. Crude, but the right measure. It becomes finer when the value to your business and the complexity of the matter come in. That is more complicated than counting hours. But it has an advantage that time billing lacks. With a fixed fee, both sides know from the outset what is owed and what is paid.

That it works without time billing is no outsider position. One of the most profitable law firms in the world has put it in its own billing terms:

The same terms record that the firm furnishes no breakdowns of individual lawyers and hours. Whoever works that way sells no hours.

In civil disputes the value-based fee has a limit that should not be concealed. According to the Federal Court of Justice, a fee is presumed to be unreasonably high where it is more than five times the statutory fees. This applies to fixed and time-based fees alike, measured against the fees the statute would have provided. Above that threshold the lawyer must prove the fee reasonable.¹⁵ A fee oriented on the value to the client can exceed that threshold. With a fixed fee, however, the lawyer can rebut the presumption with precisely that value: the importance of the matter to the client, the client’s goal, and the result achieved.¹⁶ The tension is smaller as a result, but it is not resolved.

That leaves the honest objection, and it holds. There are matters whose effort nobody can seriously predict, above all in litigation. The court, the opposing side, the evidence, the course of the proceedings, all outside anyone’s control. But even there, time billing is not the answer.

In the interest of not claiming a purity that does not exist: Since founding INN.LAW in 2015 I have billed a fixed fee by preference, though not exclusively. Where the scope cannot be seriously outlined, or where a client expressly wants it, I agree an hourly rate, record in quarter-hour increments, and round down in the client’s favor. What the courts condemn is rounding up, not the increment.

The last point of that recommendation leads to the legislature. Section 4a RVG permits contingency fees only in narrow exceptions. The provision visibly rests on the assumption that the lawyer executes what others have decided and therefore carries no entrepreneurial risk. Other jurisdictions moved past this long ago. It is the same assumption that underpinned that corporation’s fee policy, and it was wrong there too.

Munger said in 2007 that he had no solution for the hours target. For the model underneath it there is one, the fixed fee. In 1958 the American Bar Association recommended recording every hour because lawyers’ incomes rose more slowly than those of dentists.

Frequently asked questions

Is hourly billing unlawful?

No. It is lawful, and there are matters whose scope cannot be seriously outlined at the outset. The objection is about incentive and about control. The model rewards effort rather than outcome, and the client cannot see how much time the lawyer actually spends.

How is a fixed fee calculated, if not by time?

By the value to your business and the complexity of the matter. German statutory fee law (Rechtsanwaltsvergütungsgesetz, RVG) offers a first orientation by tying the fee to the value of the matter. It is crude, but it shows the principle. Effort is the floor, not the yardstick.

What about matters whose scope nobody can predict?

They exist, above all in litigation, where the effort depends on the court, the opposing side, and the course of the proceedings. The answer is not hourly billing but a fixed fee for an initial assessment, then fixed amounts for each stage of the proceedings and caps, in court not below the statutory fees, and only exceptionally a sharing of risk under Section 4a(1) sentence 1 No. 3 RVG.

Notes

1. The 1913 figures are recorded in the firm history of Smith’s own successor firm, WilmerHale, A Slice of History, 2010. ↩

2. WilmerHale, A Slice of History, 2010. ↩

3. American Bar Association, Special Committee on Economics of Law Practice, The 1958 Lawyer and His 1938 Dollar, 1958, pp. 5, 9 f., 18 f. ↩

4. Goldfarb v. Virginia State Bar, 421 U.S. 773 (1975). ↩

5. Internal firm policy, reproduced in Rubino, Above the Law, 18 August 2025. ↩

6. BVerfG, order of 15 June 2009 – 1 BvR 1342/07, para. 20. ↩

7. CJEU, judgment of 12 January 2023 – C-395/21, operative part, points 2 and 3, on Articles 4(2) and 3(1) of Directive 93/13/EEC and therefore on consumer contracts; BGH, judgment of 12 September 2024 – IX ZR 65/23, paras. 19, 29. ↩

8. BGH, judgment of 13 February 2020 – IX ZR 140/19. ↩

9. OLG Düsseldorf, judgment of 13 January 2026 – 24 U 65/22. ↩

10. BGH, judgment of 19 February 2026 – IX ZR 226/22. ↩

11. Senior Courts Costs Office, [2026] EWHC 703 (SCCO). ↩

12. BGH, judgment of 12 September 2024 – IX ZR 65/23, paras. 15 f.; BGH, judgment of 19 February 2026 – IX ZR 226/22, para. 32. ↩

13. Reported by Stephen E. Embry, ABA Law Practice Today, 2019. ↩

14. Thomson Reuters, 5 Critical Insights from the 2026 State of the US Legal Market Report, drawing on Legal Tracker data. ↩

15. BGH, judgment of 8 May 2025 – IX ZR 90/23, paras. 16 f.; BGH, judgment of 10 November 2016 – IX ZR 119/14. ↩

16. BGH, judgment of 8 May 2025 – IX ZR 90/23, para. 18. ↩

Reference: Poleacov, P. (2026). The billable hour is out of time. INN.LAW. https://inn.law/en/perspectives/billable-hours/

---