Incorporator The Journal of U.S. Formation
Start Free Quiz
Editorial 9 MIN READ

The C-corp, twenty months into the 21% rate

A field report on what the TCJA did to founder arithmetic, what the first SOI data shows, and which provisions are biting in practice

The Editors
Staff, Incorporator.org
9 min read Updated
Contents 5 sections
  1. What the first SOI data actually says
  2. The §1202 uptick is real, and it is showing up in two places
  3. §163(j) has been more disruptive than the filing population expected
  4. What the 20-month data means for the choice, and what it does not
  5. Sources

he Tax Cuts and Jobs Act has been in force for twenty months, long enough that the C-corp math we rewrote in January 2018 can now be checked against filings rather than projections. The short version: the defaults held, §1202 is doing more work than we expected, and §163(j) is the provision founders underpriced.

This is a field report. It updates our 2018 piece on the C-corp after TCJA with the first round of IRS Statistics of Income data from the 2018 tax year and what practitioners are seeing on the ground. The spine of the argument in our 2016 original, that the venture-bound founder belongs in a Delaware C-corp on day one, has only gotten stronger.

What the first SOI data actually says

The IRS Statistics of Income Division released its preliminary 2018 corporate data in the summer bulletin this year, covering Forms 1120 filed for tax years beginning in calendar 2018. That is the first full return year under the 21% rate, the repealed corporate AMT, and the new §163(j) interest limit. Preliminary SOI tables are exactly that, preliminary, and the historical revision between the preliminary bulletin and the full Corporation Complete Report has run several percentage points on aggregate income measures in prior cycles. Treat the numbers as a rough first draft.

Total corporate income tax after credits on the preliminary tables dropped roughly in line with the statutory rate cut, which is the headline that Joint Committee on Taxation staff had projected in its pre-enactment revenue tables. The drop was not proportional to the rate move because a large share of C-corp taxable income sits inside entities that had been paying something below the old 35% top rate already, through graduated brackets that bit lightly at the lower end, through domestic production activities deduction benefits under the old §199 that TCJA repealed, and through rate differentials on foreign earnings the new GILTI and FDII regimes rearranged. The net of all of that is a corporate tax take lower than pre-TCJA, higher than a naive 21/35 projection, and noisy.

Filing counts of Form 1120 in the preliminary data tick down slightly from prior years, which continues a multi-decade secular drift as operating businesses choose pass-through forms. The drift is not accelerating, and there is no evidence in the preliminary tables of a reverse migration from Subchapter S or partnership back into C corporate form. The 21% rate did not, by itself, pull operating companies out of pass-through posture. The §199A deduction did most of the work of holding them there.

What the SOI tables do not yet show is how much of the C-corp base is new-issuance qualifying small business stock under §1202. That data, if it is ever extracted cleanly, sits inside individual 1040s as exclusions reported on Form 8949 and Schedule D, not on the 1120. It runs through the shareholder's return, not the corporation's.

The §1202 uptick is real, and it is showing up in two places

The IRC §1202 exclusion was not touched by TCJA. It excludes up to the greater of $10 million or ten times basis of gain on qualifying C-corp stock held more than five years, and at the 100% tier for stock issued after September 27, 2010, it runs tax-free federally. That was already the most valuable line in the Code for a successful founder in 2018, and the combination of a 21% corporate rate and an unchanged §1202 has pushed practitioners to be more rigorous about the facts that make a share qualify.

Two concrete shifts are visible in the field through the first half of 2019.

The first is at the formation stage. Corporate-formation counsel at venture-oriented firms are now routinely papering a "QSBS memo" into the formation binder, documenting the $50 million aggregate gross assets test as of the relevant issuance date, the active-business requirement under §1202(c)(2), and the excluded-industry screen under §1202(e)(3). That screen is where most disputes will live. The excluded list strikes "services" broadly, including health, law, accounting, consulting, financial services, and any business whose principal asset is the reputation or skill of its employees. A software-enabled services company, a telemedicine platform, a fintech that looks operational but sits close to "financial services" as a matter of classification, all now get a QSBS opinion or a memo to file at incorporation rather than at exit.

The second is at exit. The IRS issued informal guidance in 2019 on the trade-or-business character of certain platform companies in the excluded-industry gray zones, though nothing that rises to a revenue ruling or regulation on point. Practitioners are preparing protective disclosures on Form 8275 for the 2018 tax year where the §1202 exclusion is being claimed on a sale of stock in a company whose industry classification is defensible but not obvious. Chief Counsel Advice memoranda released through the year have nibbled at the edges, and the case law remains thin. A founder with a clean QSBS exit in hand should file; a founder in a gray industry should paper the position.

On the tax-preparation side, there is no longer a meaningful push from pass-through-oriented shops to convert QSBS-eligible C-corps into LLCs to chase the §199A deduction. The conversion costs the five-year holding-period clock, the five-year clock restarts on a §368(a)(1)(F) or §351 reorganization in most postures, and §199A sunsets on December 31, 2025 anyway. A founder who was going to hit a §1202 exit in 2023 or 2024 is not giving that up for a deduction that expires in 2026.

§163(j) has been more disruptive than the filing population expected

The §163(j) interest-expense limitation was the provision we flagged in 2018 as a sleeper and it has woken up. The rule caps net business interest expense at 30% of adjusted taxable income for any taxpayer other than a small business under the §163(j)(3) average-gross-receipts exception, which was indexed and sits at roughly $26 million for tax years beginning in 2019 under Rev. Proc. 2018-57.

Three consequences have landed.

Any leveraged buyout structure that closed before TCJA and modeled interest deductibility at the old, essentially uncapped rule is now running below the cap for the entire post-2017 holding period. Sponsor counsel have been modeling the disallowed interest as a carryforward under §163(j)(2) since January 2018, but the first full year of cash effects showed up on 2018 returns filed in 2019. The cash tax hit across the leveraged portfolio segment is larger than sponsors expected and smaller than the statutory language reads, because adjusted taxable income under the 2018 regime adds back depreciation and amortization. That adds back is scheduled to sunset for tax years beginning after December 31, 2021, when ATI becomes an EBIT concept rather than an EBITDA concept. Sponsors building four-year holds today are pricing in that ratchet.

The second consequence is in venture debt. A late-stage growth company burning cash to scale, funded with a mix of equity and venture debt, can generate more interest expense than 30% of its adjusted taxable income, particularly if ATI is thin or negative. Before TCJA the interest was deductible when paid; now it stacks as a carryforward that survives changes of ownership only within §382 limitations. A company running through a §382 ownership change in the course of later rounds loses use of the carryforward to the extent the §382 limit does not cover it. Venture debt is still cheap; the tax shield on it is meaningfully weaker than it was.

The third consequence is in partnership structures held by C-corp investors. §163(j) applies at the partnership level first, with any excess business interest pushed out to partners and suspended at the partner level until the partnership allocates excess taxable income back. The mechanics are in §163(j)(4), and practitioners have spent the year working out edge cases on tiered partnerships and blocker-corporation structures. Treasury's proposed regulations issued in late November 2018 at REG-106089-18 run more than 400 pages in the Federal Register, and as of this writing the final regulations are still pending.

What the 20-month data means for the choice, and what it does not

The 2018 piece argued that the defaults did not change: the venture-bound founder belongs in a Delaware C-corp, the services business distributing its cash does not, and the middle case moved slightly in the C-corp direction. Twenty months in, that framing is holding.

What the 20-month data added is a slightly stronger version of the case for the venture default. The 21% rate plus intact §1202 plus repealed corporate AMT is a materially cheaper option to carry than the pre-TCJA version. At the formation stage the incremental cost of picking a C-corp "in case we raise" has dropped, and the insurance value of picking one because you actually will raise is the same as it ever was. Every venture-backed deal we have seen paper in 2019 has closed into a Delaware C-corp, and every one that started as an LLC paid the conversion toll somewhere between seed and Series A.

What it has not changed is the calculus for the cash-flow services business. A consulting firm or medical practice distributing most of its income every year still pays more total tax as a C-corp than as a pass-through, and the §199A sunset date is far enough out that the practical decision for a new firm in 2019 is to run as a pass-through and revisit in 2024. The §199A service-trade phase-out still zeroes the deduction for the highest earners in a specified service trade or business, so the most-affected partners see neither the C-corp benefit nor the pass-through benefit and pay 37% top rate plus state on every dollar. That is the population with the weakest post-TCJA position, and nothing in the first twenty months of data has changed it.

What remains genuinely unclear is the interaction between GILTI, FDII, §163(j) final regulations, and the 2021 ATI ratchet on a five-year business plan. A growth company with international IP, a modest amount of debt, and an exit horizon that straddles 2022 is modeling three moving provisions at once, and any one of them changing in final regulation form can shift the answer. The §163(j) regulations are the nearest to landing. Practitioners expect them before year-end, and the final form will determine whether the anti-avoidance rules on related-party debt are tighter, looser, or the same as proposed.

What was a projection in January 2018 is now a partial return cycle with numbers on it. The defaults held. The edge cases moved. The venture-bound founder is in the same place, slightly more so. The cash-flow pass-through owner is in the same place, with a scheduled cliff in 2025. The interesting file is the middle, and the interesting provision is §163(j), and anyone advising a C-corp through the next two years is reading the Federal Register more carefully than they were last September.

Sources

Keep reading

More from the journal.