Incorporator The Journal of U.S. Formation
Start Free Quiz
Editorial 8 MIN READ

The C-corp, revisited: the math after 21%

A headline rate that dropped fourteen points in one bill, a dividend rate that did not, and a pass-through deduction that complicates the comparison

The Editors
Staff, Incorporator.org
8 min read Updated
Contents 6 sections
  1. What the Act actually changed for a C-corp
  2. The double-tax math, after TCJA
  3. Section 199A and the counter-incentive
  4. Section 1202 is unchanged, and that matters more than anything
  5. When the C-corp makes sense now, and when it still does not
  6. Sources

n December 22, 2017, the President signed H.R. 1, and the headline federal corporate rate dropped from 35% to a flat 21% for tax years beginning after December 31, 2017. For every founder weighing a C-corp against a pass-through, the arithmetic in our 2016 piece on the vanilla default is now fourteen points out of date.

The conclusion is not that the C-corp is suddenly the right form for every business. It is that the penalty for picking it, if you pick it for the wrong reasons, is meaningfully smaller than it was last week.

What the Act actually changed for a C-corp

Three provisions do most of the work.

Section 13001 of the Act strikes the graduated corporate rate table in IRC §11(b) and replaces it with a flat 21% on all taxable income, with no brackets and no phase-outs. The rate applies to tax years beginning after December 31, 2017. For a calendar-year corporation, that is the return you will file in 2019 for the 2018 tax year. A fiscal-year corporation whose year straddles the effective date computes a blended rate under IRC §15. The personal-service corporation surtax, the old 35% flat rate for qualified PSCs under §11(b)(2), is also gone; the 21% rate applies to every C-corp including a professional corporation.

Section 12001 repeals the corporate alternative minimum tax under §55 for tax years beginning after December 31, 2017. Any AMT credit carryforwards a corporation is holding become refundable in tranches through 2021 under new §53(e). For startups that accumulated AMT credits in a prior profitable year and then used accelerated depreciation or R&D credits, those credits are now cash, not a reserve line item.

Section 13301 adds the §163(j) interest-limitation rule, capping net business interest expense at 30% of adjusted taxable income, with a $25 million gross-receipts exception for small businesses. For most operating startups this is a non-event; for leveraged portfolio companies and any C-corp with significant related-party debt, it is the first thing counsel will model. The 2016 §385 documentation regs we covered last year live inside this world now.

None of this touches the shareholder side. Qualified dividend rates remain at 0%, 15%, and 20% under §1(h)(11), and the 3.8% net investment income tax under §1411 is still on the books. A top-bracket shareholder receiving a qualified dividend still pays 23.8% federal on it, plus state.

The double-tax math, after TCJA

A dollar of corporate income, earned and distributed, now looks like this at the federal level.

The corporation pays 21% on the dollar, leaving 79 cents. The shareholder receives that 79 cents as a qualified dividend and pays 20% plus 3.8% NIIT at the top bracket, which is 23.8% of 79 cents, or about 18.8 cents. The shareholder keeps roughly 60.2 cents. The combined federal rate on fully distributed corporate income is 1 - (0.79 × 0.762), or 39.8%.

Under the prior regime, the same dollar lost 35 cents to corporate tax and then 23.8% of the remaining 65 cents to the dividend and NIIT. The shareholder kept about 49.5 cents. The combined federal rate was 50.5%.

The Act compressed the all-in federal distribution rate by roughly ten points. State taxes still stack on top, and they did not move in lockstep; a coastal company in California or New York is looking at combined all-in rates in the mid-40s now, down from the high-50s.

A pass-through owner in the top bracket pays a maximum 37% federal on ordinary income under the new §1, plus state, plus self-employment tax where applicable, but only once. Before §199A, that was a clean win against a C-corp that distributes. After §199A, the comparison depends on whether the pass-through income qualifies for the new 20% deduction.

Section 199A and the counter-incentive

Section 11011 of the Act adds new §199A, which allows individuals, trusts, and estates to deduct up to 20% of qualified business income from a domestic pass-through for tax years beginning after December 31, 2017 and before January 1, 2026. The deduction is taken below the line, does not reduce adjusted gross income, and is limited above a threshold (in 2018, $157,500 for singles and $315,000 for joint filers) by a W-2 wages and qualified property test for non-service businesses and by a phase-out that zeroes the deduction entirely for specified service trades or businesses. Law firms, accounting practices, health providers, consulting shops, and investment management are on the service list; engineering and architecture are not.

For a pass-through owner below the threshold, 20% of QBI comes off the top, and the effective top federal rate on business income drops from 37% to 29.6%. For a manufacturing LLC with a large wage base, the deduction survives the phase-out and the same math applies. For a partner at a high-earning consulting firm, the deduction phases to zero and the full 37% bites.

The §199A provision sunsets December 31, 2025. The 21% corporate rate does not. That asymmetry is intentional and is the single most important timing question in the founder's choice right now. A pass-through owner who expects to operate well past 2025 is pricing in a scheduled rate snapback; a C-corp owner is not.

Section 1202 is unchanged, and that matters more than anything

The Act left IRC §1202 alone. Stock issued by a qualifying C-corporation after September 27, 2010 and held for more than five years is still eligible for 100% exclusion from federal gain on sale, up to the greater of $10 million per taxpayer per issuer or ten times the aggregate adjusted basis in the stock. The original-issuance requirement, the $50 million aggregate-gross-assets test at issuance, the active-business requirement, and the excluded-industry list (services, finance, farming, hospitality, and several others) all remain in §1202(c) through (e).

This is the load-bearing column of the post-TCJA founder's argument for a C-corp. At a 21% corporate rate, the deferred-tax cost of earning a dollar inside the entity is much lower than it was. At a 0% federal rate on the first $10 million of qualifying gain per founder, the second layer of tax vanishes at exit for the thing founders are actually building for. The spread between a C-corp optimizing for a §1202 exit and a pass-through optimizing for annual distributions has widened, not narrowed. The bill made the default stronger for the companies it was designed for, without helping the companies it was not.

Note the deadline-agnostic structure of the provision. There is no issuance-date cap on the 100% tier. A share issued in 2018 by a qualifying company, held through 2023, will still qualify for the 100% exclusion on sale in 2023 or later.

When the C-corp makes sense now, and when it still does not

The businesses that belonged in a C-corp in 2016 still belong there. Anything you expect to raise institutional money into, any holding company whose subsidiaries may transact with each other, and any company building toward a sale in five years with §1202 eligibility on the cap table, form as a Delaware C-corp on day one. The conversion cost at a Series A is a real expense, the QSBS holding-period clock resets on a conversion under Rev. Rul. 84-111, and the cheapest time to fix all of that is the first time.

The interesting change is in the middle. A profitable operating company that generates cash but does not distribute all of it, for instance a software business reinvesting heavily in headcount and product, has a better case for a C-corp than it did a month ago. If cash accumulates inside the entity and funds growth, it is taxed once at 21% and not again until a distribution event. A pass-through owner in the same posture still pays the top individual rate on every dollar of retained earnings as it is earned, 199A deduction permitting, whether or not any of it reaches the owner's bank account.

The accumulated-earnings tax under §531 still exists and still imposes a 20% penalty on unreasonable retention, but the reasonable-needs-of-the-business safe harbor is broad and has never been the governing constraint for a growing operating company.

The businesses that should not default to a C-corp are still the same set, smaller. A cash-flow services business distributing most of what it earns will still pay more total tax as a C-corp than as a pass-through, and the gap is still ten-ish points at the coastal state level. A solo operator with no exit plan and no §1202 runway does not need the structure. A professional-services firm whose owners want the money this year wants Subchapter S or a partnership.

What changed is the strength of the default for founders who are not sure. Under the old regime, picking a C-corp "in case we raise" cost you real money every year you did not. At 21% with no corporate AMT, with §1202 intact, and with a pass-through deduction that sunsets in seven years, the option value of starting as a C-corp is cheaper to carry than it has been in a generation. The right answer for a venture-bound founder was always Delaware C-corp; the wrong answer for a non-venture founder is less wrong than it used to be.

The authorized-shares trap we wrote about in 2016 is, incidentally, the same trap today. Delaware's franchise-tax notice still prints the larger of the two methods. The elective assumed-par-value method still produces the $400 minimum for a typical early-stage charter. February is still the month founders call their lawyers about a $75,000 bill. Nothing in H.R. 1 touched any of that.

The Conference Report on H.R. 1 runs to more than 1,100 pages and there is a great deal it does that this piece does not cover, including one-time transition tax on foreign earnings under §965, the new GILTI and FDII regimes under §§951A and 250, the limitation on net operating loss deductions to 80% of taxable income under amended §172, and the tightened like-kind exchange rules that now exclude personal property. Each of these will matter to some C-corp depending on its facts. None of them changes the founder's arithmetic at the scale this article is addressed to. The founder's choice has always been, and in January 2018 remains, a three-way question: what are you building, who is paying the tax, and when does the cash come out.

Sources

Keep reading

More from the journal.