This 28th article of the Unimpeachable Neutrality Series defines intellectual capitalism, traces labor-market capitalism from Smith and Marx to Coase and Baumol, separates the professions from the nurse and the lineman whom robotics will not reach, reads the employment data against its base years, and explains why the petite bourgeoisie Marx said would be swamped is coming back with a CPA license, and why the one asset that cannot be downloaded is an unimpeachably neutral method.
In January 2017, I opened a forensic accounting firm with one employee. I had been working alone for a while before that, at a firm where my partner’s attention was elsewhere and the reports still had to go out, so the habit was already formed. For nearly a decade, every recruiter who called told me the same thing: you have judgment, but you have no leverage. They were right. The only leverage I ever had was that nobody could say I had been bought, and I built a career and this series on that one asset. This year the recruiters stopped being right, and I want to explain why, because it is not really about me.
Every economic system anyone has bothered to name is an answer to one question: who owns the tools. For most of history, the answer was whoever could afford them and everyone else worked for that person by the hour. For one kind of business, the kind that sells analysis, advice, and judgment, that arrangement is ending. This 28th article of the Unimpeachable Neutrality Series says so in writing, with a date on it.
Intellectual capitalism is what you get when professional judgment, written down clearly enough that a machine can execute it, becomes capital in the strict sense: an asset that earns a return without the owner’s hours and that compounds.
I think that is one of the more profound economic changes any of us will live through. To see why, start with what capitalism is.
DISCLAIMER: The thoughts in this article are my own, written from the slightly unbalanced perch of someone who loves professional standards so much that I made the section sign (§) my business logo. I do not speak for the Appraisal Standards Board, the NACVA Standards Board, the AI and Machine Learning Commission, or the GACVA Advisory Council. I serve on all four; nothing here represents the views of any of them.
What Capitalism Actually Is
The word gets used as a compliment and an insult, and rarely as a definition, so here is one. Capitalism is the arrangement in which the tools of production are privately owned, the people who do not own them sell their time for a wage, and the difference between what that time produces and what it is paid goes to the owner. Everything else, from the stock exchange to the billable hour, is a way of doing one of those three things more efficiently.
It did not begin with wages. Adam Smith, who is quoted more often than he is read, wrote that “the work done by slaves, though it appears to cost only their maintenance, is in the end the dearest of any,” because “a person who can acquire no property can have no other interest but to eat as much and to labor as little as possible.”[1] Wage labor replaced bondage not because the owners had a change of heart but because a worker with a stake in the output is cheaper than one without; the historians who traced the money, Eric Williams and Sven Beckert, showed that slavery financed much of the capital that built the first industrial economies.[2] Keep Smith’s word in mind: interest. The whole of expert work turns on it. An expert with an interest in the outcome is what this series has always called a hired gun, the witness whose opinion follows the retainer, and an expert with none is the only kind a court should hear. This article makes Smith’s argument one step further along: the owner does better when the worker owns a piece of the result and the court does better when the expert owns nothing but the method.
Marx supplied the vocabulary the rest of us still use. The owners of the tools were the bourgeoisie. The people with only their time to sell were the proletariat. Between them sat the small operators who owned their own tools and worked for no one, the class the tradition came to call the petite bourgeoisie, and about them he made a specific prediction: they “all sink gradually into the proletariat, partly because their diminutive capital does not suffice for the scale on which Modern Industry is carried on, and is swamped in the competition with the large capitalists.”[3] For 170 years, on the whole, he was right. Hold that thought.
Smith, once more, on why the arrangement holds: “Masters are always and everywhere in a sort of tacit, but constant and uniform, combination, not to raise the wages of labor above their actual rate.”[4] That sentence is the reason unions exist.
Labor With a Better Vocabulary
Call arrangement professionals actually work inside labor-market capitalism, because the market it runs on is the market for hours. A firm buys the hours of people with credentials, bills them to the client at a multiple of what it pays, and keeps the difference. That difference is what the professions call leverage.
For a century and a half, the professions told themselves they were exempt. We did not own factories. We owned knowledge, which lived in our heads and could not be seized. Peter Drucker gave us a flattering name for it in 1959, the knowledge worker,[5] and we adopted it without noticing that the operative word was the second one. The knowledge earned only while the knower was working. The partner was the bourgeois, the associate was the proletarian with a graduate degree, and the surplus was called leverage. Leverage has a cost that never shows up on the invoice; every rented hour belongs to someone who knows which side is paying, and neutrality that has to pass through 40 people arrives a little bent.
Ronald Coase explained in 1937 why firms exist at all: sometimes it is cheaper to organize work inside an organization than to buy it on the market and “a firm will tend to expand until the costs of organizing an extra transaction within the firm become equal to the costs of carrying out the same transaction by means of an exchange on the open market.”[6] The consequence follows directly. When the market price of a task falls, the firm shrinks. It does not fail. It just has less reason to exist.
The solo practitioner, under that arrangement, was Marx’s petite bourgeois with a CPA license; the small operator who owned his own tools, refused to become capital, and paid for the refusal exactly as predicted. I had judgment and no leverage. Marx said my diminutive capital would not suffice for the scale of modern industry. For a decade he was not wrong.
Two Kinds of Labor
Robots replace muscle. This article is not about muscle.
The International Federation of Robotics counts 4.66 million industrial robots at work worldwide and 542,000 installed in 2024 alone.[7] That is what most people mean by automation, and the robot on the assembly line is capital in the old sense: owned by the factory, operated by wage labor, accumulating to the owner exactly as Marx described. Robotics extends labor-market capitalism. It does not end it.
Nor does it reach everywhere. Hans Moravec observed in 1988 that abstract reasoning is easy for machines while the things a one-year-old does without effort, moving and perceiving in a messy world, are brutally hard.[8] The Bureau of Labor Statistics projects home health and personal care aides to add more jobs than any other occupation through 2035, some 847,000, with nurses, line installers, and plumbers growing alongside them.[9] William Baumol explained 60 years ago why that work gets relatively more expensive rather than less: when productivity rises in one sector and cannot rise in another, wages in the stagnant sector rise anyway to keep the workers; and he named health care as the leading case.[10] The nurse is not becoming capital. The nurse is becoming more valuable labor.
That is where the unions will stay, too. Membership is 10% nationally, 17.8% in utilities, and 1% in finance and insurance.[11] Smith’s combination of masters is met by a combination of workers where the work is physical and necessary, and by nothing where the work is analysis; the analysis professions never unionized because the firm gave the associate a ladder instead of a picket line.
So, to be precise about scope, I am writing about firms that sell analysis, advice, and judgment. Accounting, valuation, forensic economics, tax, advisory, and the litigation-support corner of law. In those firms, the tool of production was never a machine. It was the hours of other people with credentials and that is the one kind of capital that just became cheaper than the wage it replaced.
Then the Tools Got Cheap
Artificial intelligence (AI) did not make knowledge workers obsolete. It made the tools of knowledge work, the part that used to require renting other people’s hours, available to any individual for the price of a subscription. The associate who used to sit in the next office now sits in a browser tab, never asks for a day off, and does not laugh at my jokes.
Keynes saw the outline in 1930 and named it “technological unemployment,” meaning “unemployment due to our discovery of means of economizing the use of labor outrunning the pace at which we can find new uses for labor.”[12] Acemoglu and Restrepo gave it its modern form: automation displaces labor from old tasks, and wages depend on whether new tasks arrive fast enough to bring it back.[13] In the analysis professions, the displacement is arriving first for the entry-level task and the spread the firm collected on that task moves toward the person whose judgment the tool is executing.
The Landlord Problem
I do not own the tools. I rent them.
The engine that runs my methodology belongs to one of a handful of companies with market values larger than most countries, on a subscription that can be repriced or shut off at the landlord’s discretion. On that reading, labor-market capitalism is being replaced by platform capitalism, with a new bourgeoisie owning a new tool and everyone else renting access. Marx would recognize it immediately.
Here is the honest answer. There are two assets in the new arrangement, the engine and the methodology, and the return goes to whichever is scarcer. The engine is getting less scarce every quarter; several vendors, interchangeable models, prices disciplined by competition, and no stake whatsoever in how my case comes out. The methodology is not, because it is the accumulated judgment of one practitioner and no vendor sells that. So long as the engine stays a commodity, the spread lands on the practitioner. If the engines consolidate, or the vendors start bundling the methodology, it lands on the landlord and this article is wrong. That question is open and the rest of the argument rests on it.
One more piece of honesty. The share of business income going to labor has been falling since about 2000 and hit its lowest reading in the 79-year history of the series this past quarter.[14] That is not an AI story; it is the story AI arrived into. What is new is that the person renting the hours finally has a way to stop.
The Numbers, Read Carefully
Five figures tell the story, each measured against a pre-pandemic base year and with the caveats in the notes, because a number presented without the year it is being compared to is the oldest trick in the hired gun’s report.
Business applications hit a record 5.67 million in 2025, 60% above 2019, but applications that planned to pay wages were 501,000 against 479,000.[15] The country is forming 60% more businesses and about the same number that intend to hire anyone. Coase’s firm, shrinking, in a spreadsheet.
One-person professional businesses grew their receipts from $185 billion to $234 billion between 2019 and 2023, a 26% increase in four years.[16]
Employment of 22-to-25-year-olds in the most AI-exposed occupations is running about 19% below where it would be had it tracked their less-exposed peers; a gap that has widened at every update and does not appear among experienced workers in the same jobs.[17] The mechanism is not layoffs. Firms stopped hiring the entry-level hour; one Big Four firm plans to cut entry-level tax and audit hiring by a third over three years,[18] and the two hundred largest law firms hired the same number of first-year associates in 2025 as in 2022 through four years of revenue growth.[19]
Read together, against the base year, it is one story. The hours are leaving the balance sheet.
The Petite Bourgeoisie Returns
Here is where Marx’s prediction finally fails, 170 years late and for a reason he could not have anticipated.
He said the small operator would be swamped because his diminutive capital could not match the scale of modern industry. That was true as long as scale meant hours. It stops being true the moment the tool that provides the scale costs less than a month of an associate’s wage. A methodology written down until a machine can run it now earns on engagements the practitioner is not typing, and every correction gets written back into the instructions, so the next engagement starts from a higher floor. That loop is the compounding. If your methodology was “I know it when I see it,” there is nothing to codify. If it was fixed schedules, fixed steps, and fixed rules about what a report may and may not contain, you have been accumulating capital for years and calling it habit.
The objection writes itself: capital that can be written down is the easiest capital in the world to copy. Partly right, and the part that is right is the point. The instructions are copyable. The judgment that decides when the rule does not apply is not, and neither is the decade-and-a-half record of the same method producing the same kind of answer regardless of who was paying. That record is what this series has called unimpeachable neutrality, and it turns out to be the one asset in the new arrangement that cannot be downloaded. Under labor-market capitalism, the return went to whoever controlled the hours. Under intellectual capitalism, it goes to whoever has the best-codified and best-defended judgment, and because everyone can afford the engine, the competition is finally about the thing it should always have been about. Schumpeter called this “the perennial gale of creative destruction” and said it “is the essential fact about capitalism.”[20] The gale has reached the one asset the professions thought was sheltered.
And here is where neutrality stops being a virtue and becomes the capital itself. The tool knows who the client is; it read the engagement letter, and it is inclined to agree with whoever is typing. What makes it neutral is that the rules it applies were written before the case arrived. It annualizes partial-year earnings, meaning it scales a partial year up to a full one, because that is the firm’s rule, whether or not doing so helps this plaintiff. A staff accountant applies the rules knowing which conclusion would be welcome upstairs, and that knowledge leaks. The tool has no upstairs. Consistency of method is the whole of neutrality and it holds only as long as the practitioner writes the rules first, keeps the instructions in the work file beside the schedules they produced, checks every figure against its source himself, and says so under oath. The tool removes the hours. It leaves the accountability, which is the only arrangement under which an unimpeachably neutral expert could sign.
The Apprenticeship Problem
Here is the hole in my own thesis, and I would rather dig it than have it dug for me.
The capital I just described is 15 years of expert work, bad reports, and a screaming match with a retired justice over personal goodwill. It includes the part that cannot be taught by a machine that agrees with you, which is learning to hold a number under pressure from the people paying for it. The machine can execute that judgment. It cannot manufacture it. And firms have stopped hiring the entry-level hour, which is where every practitioner I know went to acquire the judgment in the first place. If that holds, the returning petite bourgeoisie is a closed class of people who finished their apprenticeship before 2023, and the profession is eating its seed corn.
I have three partial answers. The apprenticeship has to move from the firm to the profession, meaning the credentialing bodies and standards boards, and I say that as someone who sits on both. The machine that displaced the associate is also the best tutor an associate ever had, provided someone with judgment is standing over the shoulder. And the scarce input in 2036 is not the engine. It is the person who knows what to tell it.
Toward What?
Marx thought capitalism would produce the conditions of its own replacement. Schumpeter, who admired him and disagreed with him, thought the same thing for different reasons.[21] Smith thought a worker with a stake in his own output was the cheapest arrangement there is; that is what his sentence about slave labor being “in the end the dearest of any” was there to prove. I will not tell you which of them intellectual capitalism vindicates because I suspect the honest answer is all three. It puts the tools back in the hands of the person doing the work, which is what the critics asked for. It does so through a price drop rather than a revolution, which is what the market’s defenders would have predicted. And it leaves the nurse, the lineman, and the plumber where they were, organized and increasingly well paid, which is what Baumol would have told you.
What I will tell you is that there are two kinds of capitalism available to a firm that sells judgment in 2026, and only one has a future. Garden-variety capitalism: hire the hours, mark them up, sign at the bottom, and hope the leverage holds. Intellectual capitalism: write the methodology down until a machine can run it, keep the judgment, train the next person who will, and let the capital compound while you sleep. This is a claim about a model, not about people. The good practitioners at the firms of 40 will start writing their methodology down tonight. The firms that keep the leverage model will discover that their clients figured it out before they did.
The judgment was always the product. The organization was the delivery mechanism, and the delivery mechanism just became a subscription. What survives the transition is the one thing this series has argued for from the first installment: a method that does not know who is paying, applied by a practitioner who does and does not care. So when opposing counsel asks how many employees my firm has, the answer is still one. Marx said I would be swamped. I just own the tools now.
Zachary Meyers, CPA, CVA, has been retained in over 4,000 cases as a testifying, consulting, or joint/court-appointed expert. He has testified and been qualified as an expert specific to civil, marital, and criminal litigation. He is a nationally recognized authority in forensic accounting, business valuation, and taxation. As the Managing Member of C. Zachary Meyers, PLLC, he provides expert litigation support to courts and legal professionals, whether as a testifying expert, consulting expert, or neutral/court-appointed expert. Mr. Meyers serves as a member of the Appraisal Standards Board (ASB) of The Appraisal Foundation, where he plays a key role in the development and interpretation of the Uniform Standards of Professional Appraisal Practice (USPAP). He serves as an advisor to the NACVA Standards Board and represents the United States on the GACVA Advisory Council. He was named to Forbes’ inaugural 2025 Top 100 CPAs for Valuation Services and received Marshall University’s 2023 Alumni Award of Distinction. He is the author of the Unimpeachable Neutrality series.
Mr. Meyers can be contacted at (304) 690-2619 or by e-mail to czmcpacva@czmeyers.com.
Sources
[1] Adam Smith, Wealth of Nations (1776), Book III, ch. II (Glasgow ed. III.ii.9).
[2] Eric Williams, Capitalism and Slavery (Chapel Hill: UNC Press, 1944), ch. 3; Sven Beckert, Empire of Cotton: A Global History (New York: Knopf, 2014), chs. 2–3.
[3] Marx and Engels, Manifesto of the Communist Party (1848), Section I, Moore translation (1888).
[4] Smith, Wealth of Nations, Book I, ch. VIII (Glasgow ed. I.viii.13).
[5] Peter F. Drucker, Landmarks of Tomorrow (New York: Harper, 1959).
[6] R. H. Coase, “The Nature of the Firm,” Economica n.s. 4, no. 16 (1937): 386–405, at 395.
[7] International Federation of Robotics, World Robotics 2025, press release, September 25, 2025 (542,076 installations in 2024; operational stock 4,664,000).
[8] Hans Moravec, Mind Children (Cambridge, MA: Harvard University Press, 1988), ch. 1.
[9] U.S. Bureau of Labor Statistics, Occupational Outlook Handbook, 2025–35 projections: home health and personal care aides (+847,300); registered nurses (+194,700); electrical power-line installers (+10%); plumbers, pipefitters, and steamfitters (+7%).
[10] William J. Baumol, “Macroeconomics of Unbalanced Growth,” American Economic Review 57, no. 3 (1967): 415–426; Baumol, The Cost Disease (New Haven: Yale University Press, 2012).
[11] BLS, “Union Members — 2025,” USDL-26-0229, February 18, 2026, Tables 1 and 3 (2025 figures are eleven-month averages excluding October).
[12] J. M. Keynes, “Economic Possibilities for our Grandchildren” (1930), in Essays in Persuasion (London: Macmillan, 1931), 358–373.
[13] Daron Acemoglu and Pascual Restrepo, “Automation and New Tasks,” Journal of Economic Perspectives 33, no. 2 (2019): 3–30.
[14] BLS, Nonfarm Business Sector: Labor Share, series PRS85006173 (index 2017=100), via FRED; 2026 Q2 = 93.4, series minimum since 1947; 113.6 in 2001 Q1.
[15] U.S. Census Bureau, Business Formation Statistics, release of September 11, 2026; annual totals computed from the not-seasonally-adjusted monthly file.
[16] U.S. Census Bureau, Non-employer Statistics, 2019 and 2023, NAICS 54 (3.77 million firms and $185 billion in 2019; 4.08 million and $234 billion in 2023). Census applies noise infusion to receipts data.
[17] Brynjolfsson, Chandar, and Chen, “Canaries in the Coal Mine? Six-Month Update,” Stanford Digital Economy Laboratory, August 12, 2026 (ADP payroll data through June 2026). The authors note the decline becomes significant only after 2024 and that interest rates and post-pandemic over hiring contribute.
[18] PwC US hiring plan as reported by Business Insider, August 21, 2025 (tax and assurance associate hiring from 3,242 in 2025 to 2,197 in 2028); PwC confirmed reduced campus hiring goals. Secondary source.
[19] SurePoint Law School Hiring Report 2026 (AmLaw 200 first-year hires: 7,489 in 2022; 7,426 in 2025). Secondary source.
[20] Joseph A. Schumpeter, Capitalism, Socialism and Democracy (New York: Harper, 1942), ch. VII, 83.
[21] Schumpeter, Capitalism, Socialism and Democracy, Parts I and II.
