Law Firm Partner Compensation: How Small Firms Split the Money

Most of what is published about law firm partner compensation describes firms with hundreds of lawyers, a compensation committee and a points system with a written appeals process. Almost none of it describes the situation the great majority of American law firm owners are actually in, which is two or three or five lawyers deciding, once a year, how to divide a number that is smaller than any of them would like. That decision is not really about money. It is about what the firm has agreed to pay for, what the owners are allowed to agree to under their own conduct rules, and how the money is permitted to move once the division is settled. Those three things are separable, and most small firm compensation arguments are arguments about one of them conducted in the language of another.

What partner compensation actually decides, and why it is three questions rather than one

A partner compensation plan looks like a single document, so firms treat it as a single decision. It is not. It sits on top of two other decisions that are usually older, usually undocumented, and usually the real source of the disagreement.

The first is ownership. Who holds an interest in the firm, in what proportion, and what did they pay or risk to hold it. Ownership decides who votes, who signs the lease and the line of credit, who absorbs a bad year, and who receives value if the firm is ever sold or wound up. It does not, by itself, decide annual pay, although in many small firms it is used as a proxy for it.

The second is compensation: what each owner receives out of this year’s profit, and on what basis. This is the part firms write down and argue about. It is downstream of ownership because the pool being divided is whatever remains after the firm has paid everyone who is not an owner and everything that is not a person.

The third is mechanics: how the money physically reaches the partner, in what instalments, with what tax treatment and what withholding. This is the part firms most often get wrong without noticing, because it looks like an accounting detail and is in fact the reason a partner’s take home does not match the number in the plan.

Separating the three is not a drafting nicety. It is the thing that lets a firm say yes to one question without conceding the others. A senior associate can be offered a share of profit without being offered ownership. An owner can take a smaller draw in a tight quarter without renegotiating their percentage. A partner who originates heavily can be paid for it without acquiring a larger claim on the firm’s eventual sale value. Firms that hold the three together in one conversation find that every question becomes a referendum on every other question, which is why those conversations go badly and get postponed for another year.

The three decisions behind one compensation plan

There is a practical test for whether your firm has separated them. Ask whether a partner could be paid more this year without owning more of the firm next year, and whether a partner could reduce their draw for two quarters without anyone treating it as a reduction in their standing. If the answer to either is no, the three decisions are fused, and the compensation plan is carrying weight it was never designed to carry. Our guide to law firm profitability covers how the pool itself is produced, which is a different problem from how it is divided.

Equity partner, income partner, and the line that actually separates them

The vocabulary is inconsistent across firms, which is part of why the question keeps being asked. The useful distinction is not the word on the business card. It is whether the person holds an ownership interest in the entity.

An equity partner holds an interest. They share in profit and, in a bad year, in loss. They usually have a capital account, usually have voting rights proportionate to their interest or as set out in the agreement, and are usually personally exposed to the firm’s obligations through guarantees on the lease and the credit line. Their income is a share of what is left, which means it is not knowable until the year is substantially over.

A non-equity partner, sometimes called an income partner or a salaried partner, generally does not hold an interest. They are compensated, often generously, but they are compensated as a cost of the firm rather than out of its residual profit. They typically do not vote on partnership matters, do not contribute capital, and do not share in sale value. For most purposes, including the tax mechanics discussed further down, they are an employee of the firm.

That is the line. Everything else, including the title itself, is a decision about how the firm presents its people, and it is a decision with its own constraints: a firm’s communications about its lawyers are subject to the rule against false or misleading statements, so a title should describe the arrangement rather than dress it up.

Two practical consequences follow. The first is that promoting an associate to non-equity partner changes the firm’s cost structure and not its ownership, which is why it is a reversible decision in a way that admitting an equity partner is not. The second is that a non-equity tier is a genuine answer to the question of how to retain a lawyer the firm is not ready to make an owner, provided the firm is honest with itself that it has deferred the ownership question rather than answered it.

A practice management column in a state bar publication, describing a real insurance defense firm’s arrangement, recorded a structure of exactly this shape: non-equity partners who “receive a salary and a performance bonus based upon working-attorney-collected-received above an established threshold and a delegation bonus,” with the firm’s non-equity partners all paid salaries above $100,000. The detail worth noticing is the delegation bonus. The firm was paying its non-equity partners to hand work down rather than hoard it, which is a compensation design decision rather than an accounting one.

Why the published lawyer salary figures do not describe a partner at all

Every owner who searches for partner compensation data runs into the same wall of salary aggregator pages, and most of them are built on the same federal source. It is worth knowing exactly what that source measures, because it excludes the population the question is about.

The Bureau of Labor Statistics Occupational Employment and Wage Statistics programme publishes national wage estimates for lawyers. In its own published documentation, under the question of how employees are defined, the programme states: “Employees are all part-time and full-time workers who are paid a wage or salary. The survey does not cover the self-employed, owners and partners in unincorporated firms, household workers, or unpaid family workers.” Asked separately whether its estimates include the self-employed, the programme answers: “No.”

So the figures below are real, current and useful, and they describe employed lawyers. They do not describe an equity partner in a law firm partnership, because that person is definitionally outside the survey frame. Any page that presents a national lawyer wage figure as an answer to what partners earn is answering a different question from the one that was asked.

Lawyer wages, May 2025, and who is missing from them

This is not a complaint about the data. It is the single most useful thing a firm owner can know before walking into a partner compensation discussion, because it tells you that the external benchmark everyone in the room is half-remembering does not apply to anybody in the room. The relevant comparison for an equity partner is not a national wage percentile. It is what this firm produced this year, what it cost to produce, and what the partners agreed in advance about dividing the remainder. Our post on attorney compensation works through the salary cluster in detail, including what those percentiles do and do not cover for employed lawyers.

The four systems firms actually use, and the behaviour each one buys

Strip away the vocabulary and there are four basic ways to divide a pool, plus the hybrids that most firms end up with. Each one is internally coherent. Each one pays for something specific, and will get more of whatever it pays for.

Equal or ownership share. The pool is divided by ownership percentage, which in many two and three partner firms is equal. It is the simplest system to administer and the only one that requires no measurement at all. It works where the partners genuinely contribute comparably, or where they have decided that the stability of not measuring is worth more than the precision of measuring. It fails, loudly, the first year one partner’s contribution diverges and neither of them wants to say so.

Seniority or lockstep. Shares move with years of ownership on a published ladder. It rewards tenure and institutional contribution, suppresses internal competition for clients, and is the system most associated with firms that want work to be handed around freely. Its weakness is that it is slow to recognise a large change in contribution in either direction.

Formula, sometimes called eat what you kill. Shares are computed from measured inputs, usually some combination of personal collections and origination. It is transparent and self-executing, and it is also the system that most reliably produces the behaviour it measures. A long-running practice management column in a state bar journal put the objection directly: eat-what-you-kill systems “might be appropriate for lone-ranger firms. However, they aren’t appropriate for firms wanting to build and create a team-based practice, since such systems typically reinforce ‘lone ranger’ behavior resulting in a ‘me first vs. firm first’ orientation.”

Subjective or committee. A person or small group reviews each partner’s year across many dimensions and assigns a share. The same column describes subjective plans as those that “review the performance of each partner and subjectively determine a relative value for each partner,” against formula plans that “use a formula to assign value to various criteria to determine compensation,” and concludes that “subjective or combination plans are most appropriate for firms desiring to build and reinforce a team-based practice. They focus on the long as well as short term and incorporate all or most compensation criteria.” The cost is that it requires trust in the decision maker, which is a thin resource in a firm already arguing about money.

It is worth being precise about what “pays for” means here, because it is not a metaphor. A compensation system is a standing instruction to every partner about where to put the next hour. Under an equal split the instruction is silent, which is fine while the partners read the silence the same way. Under a formula the instruction is explicit and arrives weekly, because each partner can see their own number moving. Under a subjective plan the instruction is whatever the partners believe the reviewer will notice at year end, which is why a subjective plan with no published criteria is the least predictable of the four rather than the most flexible.

The second thing to hold onto is that these systems are not ranked. There is no evidence that any of them produces a better firm in the abstract, and the published practice management commentary is careful to frame the choice as a fit question rather than a quality one. A two partner firm with similar practices and similar books may run on an equal split for twenty years without strain. The same split in a firm where one partner is building a referral network and the other is clearing a high volume docket will be renegotiated inside three years, not because anyone acted badly but because the system stopped describing the firm.

Most firms above two or three partners end up with a hybrid, because the pure systems each fail in a predictable direction and a hybrid lets a firm choose how much of each failure it is willing to accept. The design question is not which system is correct. It is which behaviour the firm needs more of next year, and how much measurement the partners will tolerate to get it.

One design note that applies to all of them. Whatever the system, it should produce a number before the year starts, not after. A plan that can only be computed in arrears is not a plan, it is a reconciliation, and partners who do not know in January what they are being paid for will optimise for the wrong thing all year. Our guide to running a law firm treats the compensation plan as one of the systems that has to be written down rather than carried in someone’s head.

One firm, one year, three systems and three different answers

The fastest way to see that the system is the decision is to hold the firm constant and change only the method. The arithmetic below is illustrative. Its inputs are stated so you can substitute your own, and no part of it is a claim about any real firm or any expected result.

Take a two partner firm. Collections for the year are $1,400,000 and overhead, including every non-owner salary, is $800,000, leaving $600,000 available for partner compensation. Partner A personally collected $800,000 of the firm’s work and is credited with originating $1,050,000 of it. Partner B personally collected $600,000 and is credited with originating $350,000. The two hold equal ownership.

Under an equal split, each partner receives $300,000. Under a pure formula weighting personal collections and origination equally, Partner A’s share works out to about 66 percent of the pool and Partner B’s to about 34 percent, which is roughly $396,400 and $203,600. Under a hybrid that distributes 60 percent of the pool by ownership and the remaining 40 percent by that same formula, the result is roughly $338,600 and $261,400.

The same two partners, the same year, three methods

Three observations are worth drawing out, because they generalise beyond the numbers. The first is that the pool never changed. Nothing in this exercise made the firm a dollar richer or poorer; the only thing that moved was the boundary between two people. The second is that the spread between the partners moved from nothing to nearly a third of the pool, which is why the choice of system is not an administrative preference. The third is the one firms miss: the formula result depends entirely on how origination was credited, and origination credit is an allocation decision made by people, not a fact discovered in the accounting system.

That last point is where small firm compensation plans actually come apart, so it deserves its own treatment.

Origination credit, the most argued line in any small firm plan

Origination credit is the firm’s internal answer to the question of who is responsible for a client being here. It feels like a matter of record and it is almost never recorded cleanly, because clients arrive through overlapping channels: a referral to a named lawyer, a referral to the firm, a search result, a former client returning with a new matter, a matter opened by one partner and run by another.

The practice management literature published through state bar associations is unusually practical on this, and the advice is consistent. One column, responding to a two partner firm that had been compensating on ownership percentage adjusted for personal collections and wanted to add origination, cautioned about exactly the structural problem a small firm has: “With only two partners you don’t have anyone to serve in the capacity of attribution police when and if there are disagreements as to origination credit.” Its recommendation was to “share origination credit when appropriate, allocate to firm when it appears that a client came to the firm based upon firm brand or name recognition, and consider a 5-year sunset provision whereby the credit reverts to firm or responsible attorneys.”

Those three devices do most of the work. Shared credit removes the all-or-nothing quality that makes attribution worth fighting over. Firm credit acknowledges that some clients came for the firm rather than for a person, which is both true and the thing a marketing investment is supposed to produce. A sunset stops a single introduction made years ago from paying an annuity forever, which is the failure mode that turns a formula firm into a firm where nobody wants to service anybody else’s client.

There is a second design choice underneath: whether credit attaches at the client level or the matter level. A later column in the same series, advising a firm whose associates and non-equity partners could not realistically win clients outright, recommended that “a better origination measure to track in your situation might be new matter origination rather than client origination,” and added a point that cuts against the instinct to formalise everything: “I suggest that you track, and not directly compensate, origination at the non-equity partner level. Track and reward via a salary increase or discretionary bonus instead.”

Whatever a firm chooses, the rule that matters is that it has to be chosen before the year it governs, in writing, and applied to the next client who walks in rather than to the one currently in dispute. A credit policy written to resolve a live argument will be read as a judgment about the people arguing, and it will not survive.

How the money actually reaches a partner, and why it is not a paycheck

This is the part that surprises newly admitted partners and occasionally surprises the firm admitting them. A partner’s compensation does not arrive the way an associate’s does, and the difference is not cosmetic.

The Internal Revenue Service states the position plainly on its partnerships page: a partnership “does not pay income tax,” it “passes through profits or losses to its partners,” and “each partner reports their share of the partnership’s income or loss on their personal tax return.” On employment status the same page is equally direct: “Partners are not employees and shouldn’t be issued a Form W-2.”

The distinction is not academic for the partner receiving the money. An employee’s salary arrives net of withholding, on a schedule the employer controls, with the employer carrying half of the payroll tax and remitting the rest. A partner’s money arrives gross, on a schedule the partnership sets, with nothing withheld and nothing remitted on the partner’s behalf. Two people can be paid the same headline figure and have materially different amounts of it available to spend, and the gap is not a matter of negotiation.

Three instruments do the work that a salary does for an employee. A draw is a withdrawal against the partner’s expected share of the year’s earnings; Publication 541 describes “a withdrawal by a partner in anticipation of the current year’s earnings” as a distribution. A guaranteed payment is different in kind: the same publication defines guaranteed payments as “those made by a partnership to a partner that are determined without regard to the partnership’s income,” treated for certain purposes “as if they were made to a person who is not a partner,” under Internal Revenue Code section 707(c). And the distributive share itself is reported to the partner on Schedule K-1 at the end of the year.

The practical consequence is a timing and cash flow problem that every new partner meets once. Nothing is withheld. Income tax and self-employment tax are the partner’s own responsibility, paid in estimated instalments, out of money the firm has already handed over. A partner who treats a draw as take home pay will be short in April. Firms that handle this well set the draw deliberately below the expected share, hold the difference, and true up after the year closes, which also protects the firm if the year disappoints.

The self-employment exposure is the other half of it. The IRS states that “the self-employment tax rate is 15.3%” and that “the rate consists of two parts: 12.4% for social security (old-age, survivors, and disability insurance) and 2.9% for Medicare (hospital insurance),” with an additional 0.9 percent Medicare tax applying above filing-status thresholds of $200,000 for single filers, $250,000 for married filing jointly and $125,000 for married filing separately. For an employee, half of the equivalent payroll tax is carried by the employer. For a partner there is no employer to carry it, so the whole of it is the partner’s to pay, which is a real difference in take home on the same headline number.

A plan that does not say which of these instruments is being used for which part of a partner’s compensation will be interpreted differently by the partner, the bookkeeper and the accountant, and all three will be working in good faith. Saying it explicitly costs one sentence and removes an entire category of year end surprise.

Why the entity you chose changes the compensation conversation

Many small firms are not general partnerships at all. They are professional corporations, professional limited liability companies or limited liability partnerships, and some of those have elected to be taxed as S corporations. That election changes the mechanics above substantially, and a compensation plan written for one entity does not transfer cleanly to the other.

Where the firm is taxed as a partnership, the owner is not an employee, takes draws and guaranteed payments, receives a K-1, and pays self-employment tax on earnings from the business. Where the firm has elected S corporation treatment, the owner who works in the business is an employee of it, and the IRS position on compensation is explicit: “S corporations must pay reasonable compensation to a shareholder-employee in return for services that the employee provides to the corporation before non-wage distributions may be made to the shareholder-employee,” and “distributions and other payments by an S corporation to a corporate officer must be treated as wages to the extent the amounts are reasonable compensation for services rendered to the corporation.”

How a dollar of firm profit reaches an owner

Why this belongs in a compensation discussion rather than an accounting one: the S corporation route forces the firm to answer a question the partnership route lets it defer, which is what each working owner’s services are actually worth as a wage, separately from what their ownership is worth as a return. The IRS lists the factors courts have recognised in testing that figure, among them training and experience, duties and responsibilities, time and effort devoted to the business, dividend history, payments to non-shareholder employees, comparable compensation in similar businesses, and the existence of compensation agreements and formulas.

Read that list again as a compensation designer rather than as a taxpayer. It is a reasonable description of the inputs a sensible partner compensation plan would weigh anyway. A firm that has done the work of setting defensible reasonable compensation has also, incidentally, done most of the work of separating the labour return from the ownership return, which is the distinction every compensation system is ultimately trying to draw.

Capital, the buy-in, and what a capital account is actually for

Compensation divides this year’s profit. Capital is a different question entirely: what each owner has put into the firm and left there, so the firm can meet payroll in a slow month, carry work in progress, and fund growth without borrowing on every occasion.

Firms conflate the two constantly, usually by admitting a partner to a share of profit without asking for capital, and then discovering that the new partner’s share of the distribution is funded by the old partners’ working capital. A practice management column answering a sole equity owner of a 25 attorney litigation boutique set out the counter-position: new partners should be expected to have “skin in the game,” and the same column observed that “capital contributions are all over the board – ranging from zero to $100,000,” with the adviser noting that “in firm’s your size I often see capital contributions ranging from $25,000 to $50,000.” That range is an observation about firms of that size rather than a standard, and a two or four lawyer firm should expect its own numbers to differ.

The funding mechanism matters as much as the amount. The same adviser suggested allowing “new partners a reasonable period of time to fund their capital accounts – say one or two years via a capital note,” and in an earlier column describing a four partner firm where “each of the four partners contributed $25,000 to their capital accounts,” suggested a longer runway of up to five years or help arranging bank terms. A capital requirement that a newly admitted partner cannot meet is a capital requirement that will be quietly waived, which leaves the firm undercapitalised and the plan discredited at the same time.

It also helps to be clear about what a capital account is not. It is not a measure of what a partner is worth to the firm, and it is not a prepayment of next year’s compensation. It is the firm’s own working capital, contributed by its owners instead of borrowed from a bank, and it is on the balance sheet rather than in the profit and loss account. A firm that treats capital as a toll for admission will set the number by reference to what feels like a serious commitment. A firm that treats it as working capital will set the number by reference to how many weeks of overhead it needs to be able to cover without drawing on a credit line, which is a question with an actual answer.

Three questions are worth settling in writing at the same moment, because they are far harder to settle later: what the contribution is, whether it is tied to the ownership percentage and the expected income share, and what happens to it when the partner leaves. A capital account with no stated exit mechanism is a dispute waiting for a trigger.

Our guide to starting a law firm covers the entity and banking decisions that sit underneath all of this.

The rules that constrain what your plan is allowed to say

A partner compensation plan is a private agreement between lawyers, but it is not unconstrained. Four provisions of the conduct rules reach into it directly. The text quoted here is Washington’s adopted version, chosen because it verifies as published rule text; your own jurisdiction’s adopted language is what governs you and may differ.

The first is the prohibition on sharing fees with a nonlawyer. The rule opens: “A lawyer or law firm shall not share legal fees with a nonlawyer,” and then names its exceptions. Those exceptions matter to small firms in one specific way, because one of them permits precisely what a firm usually wants: “a lawyer or law firm may include nonlawyer employees in a compensation or retirement plan, even though the plan is based in whole or in part on a profit-sharing arrangement.” A practice administrator can be in a profit-sharing plan. The administrator cannot be given a share of the fee from a matter.

The second is the prohibition on partnership with a nonlawyer: “A lawyer shall not form a partnership with a nonlawyer if any of the activities of the partnership consist of the practice of law.” The corporate analogue is stricter still, barring a lawyer from practising in the form of a professional corporation for profit if “a nonlawyer owns any interest therein,” or is “a corporate director or officer (other than as secretary or treasurer),” or “has the right to direct or control the professional judgment of a lawyer.”

The third is the restriction on agreements that limit the right to practise. The rule provides that a lawyer shall not participate in offering or making “a partnership, shareholders, operating, employment, or other similar type of agreement that restricts the rights of a lawyer or an LLLT to practice after termination of the relationship, except an agreement concerning benefits upon retirement.” Its comment explains the reasoning: such an agreement “not only limits their professional autonomy but also limits the freedom of clients to choose a lawyer.” This is the provision that catches firms importing a non-compete clause from a commercial template into a partnership agreement.

The fourth applies when the money is divided with a lawyer outside the firm, which is a fee division rather than compensation but is often drafted in the same document. That rule permits a division between lawyers who are not in the same firm “only if” the division “is in proportion to the services provided by each lawyer or each lawyer assumes joint responsibility for the representation” and “the client agrees to the arrangement, including the share each lawyer will receive, and the agreement is confirmed in writing.”

None of this is exotic. The point is that a compensation plan is one of the few firm documents where a drafting habit borrowed from general commercial practice can produce a provision the firm is not permitted to make, and the two most common imports, a profit share for a key non-lawyer employee and a restrictive covenant, are the two that sit closest to the line.

Writing the plan, a sequence that works in a small firm

The order matters more than the sophistication. Firms that start with the formula spend months on weightings and never resolve the ownership question underneath. Firms that start with ownership find that the formula question becomes much smaller.

Settle ownership first, in percentages, in writing, with a stated method for changing it. Then establish the pool: what is the firm’s profit before owner compensation, computed the same way every year, and what reserve comes off the top before anyone is paid. Then choose the system, by asking which behaviour the firm is short of rather than which method is fairest in the abstract. Then write the credit rules, including origination, before they are needed. Then set the mechanics: draw levels, timing, the true-up, and who signs off. Then agree the review cadence and the exit terms, because a plan without a stated way to change it will be changed anyway, acrimoniously.

The order to settle a partner compensation plan in

Two habits separate plans that survive from plans that get rewritten in anger. The first is that the plan is applied prospectively and only prospectively. A change agreed in October to take effect in January is a policy; the same change agreed in October to apply to the year now ending is a redistribution, and it will be read as one. The second is that the plan states what happens in a bad year, before there is one. Most compensation documents describe how to divide a surplus and are silent on how to absorb a shortfall, which means the shortfall gets absorbed by whoever is least willing to argue.

Finally, the plan is worth testing against the firm’s own numbers before it is adopted. Run the proposed system over last year’s actuals and look at the result. If it produces a number that one partner would have refused, better to find that out in a modelling spreadsheet than in the first distribution. Our material on attorney billing and on lawyer hourly rates covers the collection and rate data a plan of this kind depends on, since a compensation formula is only as good as the figures feeding it.

Frequently asked questions