If one case defines modern S corporation reasonable compensation enforcement, it is David E. Watson, P.C. v. United States, 668 F.3d 1008 (8th Cir. 2012). It is the case the IRS cites, the case practitioners cite, and the fact pattern that turned “low salary, high distributions” from a quiet habit into a known audit target. It is also widely misquoted, so it is worth being precise about what happened and what the court actually held.
The facts
David Watson was a CPA. His professional corporation, David E. Watson, P.C., was a partner in an established Iowa accounting firm, and Watson worked full time in the practice. His corporation paid him a salary of $24,000 per year. At the same time, he received $203,651 in distributions. The government’s position was simple: a substantial part of those distributions was really payment for Watson’s services, and payment for services is wages subject to FICA, whatever the corporation chooses to call it.
The district court agreed. Relying on market evidence of what comparable accounting professionals earned, it upheld the recharacterization of $91,044 as wages. Watson appealed.
The holding
The Eighth Circuit affirmed, and stated the controlling principle plainly:
Turning to the present case, we conclude the district court properly determined “that the characterization of funds disbursed by an S corporation to its employees or shareholders turns on an analysis of whether the payments at issue were made as remuneration for services performed.” … [B]ecause the district court applied the correct legal standard, we affirm its determination on Watson's FICA wages.
The test is substance: were the payments remuneration for services performed? Board minutes, distribution labels, and the owner’s intent do not settle the question. What the services were worth in the market does.
Why the government won
The detail that matters for every owner reading this: the government did not win with a ratio or a rule of thumb. It won with data. The figure the court accepted, $91,044, was built from evidence of what comparable professionals in comparable firms actually earned. Watson’s $24,000 could not survive contact with that evidence, because it was not derived from anything. It was a round number chosen by the person receiving it.
That asymmetry decides these cases. One side shows up with market comparables; the other side shows up with a number and no file behind it. The Treasury regulation has defined the standard this way for decades: reasonable compensation is the amount “ordinarily paid for like services by like enterprises under like circumstances” (Treas. Reg. § 1.162-7(b)(3)). Watson is what that regulation looks like when it is enforced.
What Watson does not hold
Three misreadings circulate. None survives reading the opinion.
- Watson does not create a ratio. The court did not say salary must be some percentage of distributions or profit. The $91,044 figure came from market data for Watson’s profession, seniority, and firm, not from a split.
- Watson does not bless any minimum. Nothing in the opinion says $24,000 is too low in general or that some other figure is safe in general. The analysis is facts and circumstances, and the IRS says the same: “There are no specific guidelines for reasonable compensation in the Code or the Regulations” (IRS Fact Sheet FS-2008-25, p.2).
- Watson is not about greedy outliers only. Watson was a credentialed professional in a legitimate, profitable practice, with a salary that was not zero. The case turned on the gap between what he paid himself and what the market said his services were worth.
The lesson
Watson reads like a documentation case because it is one. The way to be on the right side of it is to do, in advance, what the government did at trial: price your actual services against occupational market wage data, reconcile the result across recognized methods, and keep the file. If your compensation number can show its work, you are litigating from the strong side of the Watson asymmetry. If it cannot, you are Watson.