Equities | Fiscal Policy | US
On the 22 December 2017, President Trump signed into law the Tax Cuts and Jobs Act (TCJA). This ‘unified framework for comprehensive tax reform’ was the first broad overhaul of the US tax code in thirty years. Expectations were that it would have major financial consequences for firms, however, its breadth and speedy inception made exact predictions hard.
This paper, a collaboration between authors from the University of Zurich and Harvard, uses financial statements released in 2018 to quantify accurately the benefits and losses to individual firms. They do this by comparing effective tax rates (ETRs) before and after the reforms, as well as measuring changes to nonrecurring taxes. On this front, the authors find the following:
This article is only available to Macro Hive subscribers. Sign-up to receive world-class macro analysis with a daily curated newsletter, podcast, original content from award-winning researchers, cross market strategy, equity insights, trade ideas, crypto flow frameworks, academic paper summaries, explanation and analysis of market-moving events, community investor chat room, and more.
The Tax Cuts and Jobs Act – Some Win, Some Lose, Markets Got It Wrong
On the 22 December 2017, President Trump signed into law the Tax Cuts and Jobs Act (TCJA). This ‘unified framework for comprehensive tax reform’ was the first broad overhaul of the US tax code in thirty years. Expectations were that it would have major financial consequences for firms, however, its breadth and speedy inception made exact predictions hard.
This paper, a collaboration between authors from the University of Zurich and Harvard, uses financial statements released in 2018 to quantify accurately the benefits and losses to individual firms. They do this by comparing effective tax rates (ETRs) before and after the reforms, as well as measuring changes to nonrecurring taxes. On this front, the authors find the following:
- The TCJA slashes corporations’ median effective tax rates from 31.7% to 20.8% (Finding 1). Indeed, every sector experienced an effective tax rate decline. The authors emphasise, however, that this does not mean that the TCJA had benefitted every firm.
- Specifically, when revisiting the financial statements of all firms in the sample and controlling for pre-TCJA ETRs, foreign revenue and firm-level characteristics, both winners and losers are observable (Chart 1). Standout winners are in medical equipment, measuring and control equipment, and electronic equipment industries.
- Other winners who benefited from a recurring decline in their effective tax rates:
- Coal, utilities, restaurants, hotels, petroleum, natural gas, electrical equipment, computers, banking, insurance, business services, business supplies, aircraft, beer, liquor, transportation, steel works, chemicals and healthcare.
- Other winners who benefited from a recurring decline in their effective tax rates:
- Importantly, not every firm was a winner. Depending on the tax measure, 15-30% of firms experienced a tax increase (Finding 2). These were more likely to be smaller firms in printing and publishing, rubber and plastic products, or real estate.
- Other sectors that lost out because of Trump’s TCJA:
- Defence, tobacco, recreation, textiles, shipping containers, communication (incl. Facebook, Netflix and Google), construction, trading, apparel and retail.
- Other sectors that lost out because of Trump’s TCJA:
- The impact of nonrecurring taxes on firms was also unequally distributed. Approximately 20% of firms had either nonrecurring tax benefits or expenses that exceeded 3% of their total assets (Finding 3). The winners were in transportation, beer and liquor, and entertainment, while the losers in defence, electrical equipment, and trading industries.
Chart 1: There are both big winners and big loser from the TCJA
Parallel to the above investigation, the authors analyse how effectively market participants impounded TCJA information within stock prices during the legislative process. Ex-ante proxies, such as the effective tax rate before reform or foreign cash holdings, are compared to ex-post outcomes. They find this:
- Market prices reflected proxies, but these proxies account for just half of the actual TCJA impacts (Finding 4). Otherwise stated, market participants found it hard to anticipate fully the consequences during the passage of the TCJA.
- The actual impact of the TCJA was highly relevant to firm value, but much of that impact was only impounded into prices as it became visible in firms’ financial statements in early 2018 (Finding 5). This, the authors say, indicates ‘that investors find it hard to predict large and immediate changes to company cash flows due to unfamiliar events’ (Finding 6).
The Data and an Outline of the Paper
The authors collect financial statements of Russell 3000 index firms with a stock price of at least $5 on 1 November 2017 in order to estimate the annual tax implication of the TCJA. These implications, the authors argue, can be subdivided into two distinct quantifiable effects:
- A recurring effect – This comes in the form of changes to, for example, the corporate tax rate, capex expensing and limitation on interest deductibility. These all affect a firm’s effective tax rate. Computing this effect is simply the change in the ETR between the first fiscal year after the reform and the last (or last five) fiscal year(s) before the enactment. This information is available to the authors on Compustat.
- A nonrecurring effect – This has two components: (a) the remeasurement of deferred tax assets and liabilities to account for the lower statutory tax rate, and (b) the deemed repatriation tax. GAAP rules require tax changes to be recorded in financial statements of the accounting period in which they are enacted. Therefore, this effect is disclosed under ‘nonrecurring income taxes’ in 2017 financial statements. Some firms report components (a) and (b) separately. Where they do not, the paper estimates them.
Using financial statements, the authors estimate the above effects for each firm in the sample – these effects they refer to as the actual effects (ex-post). Unique to this paper, they then use regressions to analyse the factors that explain variations in these actual effects. Specifically, they look at foreign revenue, a firm’s interest deductibility curtailed, tax loss carryforwards, capital expenditure, R&D expenditure, market value of equity (log), sales growth and ROA.
Finally, the authors test stock price responses to the passage of the TCJA using proxies for the expected effect of the act’s various provisions. Comparing this ex-ante measure of expected effects with the ex-post measure of actual effects, they determine how well the market priced in the TCJA’s outcomes.
A Note on the Repatriation Tax
Pre-TCJA, US multinational enterprises were subject to a tax on any repatriated earnings. Although foreign tax credits could offset such a tax, in most instances it was not financially beneficial to transfer earnings back from abroad.
The TCJA aimed to disincentivise the indefinite deferral of foreign income. The US did this by shifting to a quasi-territorial tax system in which profits are only taxed where they are earned. In transition to this system, firms have been subject to a one-off repatriation tax at a rate of 15.5% on liquid assets and 8% on illiquid assets. A substantial inflow of foreign earnings (FRS, 2019) and increase in share buybacks followed the implementation of the one-time tax.
How Did Taxes Change?
The paper relies on two standard ETR measures, the GAAP ETR (tax expenses relative to current year pre-tax income) and the Cash ETR (percent cash taxes paid relative to current year pre-tax income). The results are as follows:
- Aggregate GAAP taxes fell from 26% to 19%. The fall was smaller for the Cash measure but still went from 24% to 20% between 2016 and 2018. The correlation between changes in GAAP and Cash is low.
- Although most firms enjoyed large tax savings on a recurring basis, about 15% (in the case of the GAAP ETR) or 30% (in the case of the Cash ETR) experienced increases.
The graph below summarises the effective tax rate changes from pre-TCJA (white) to post-TCJA (blue) across the full Russell 3000 sample. Notice the large leftward shift – firms on average pay around 16% less in recurring taxes.
Chart 2: The TJCA reduces annual tax payment
Regarding nonrecurring taxes, the authors find striking evidence that many corporations faced very large (>40%) GAAP ETRs. The degree of heterogeneity across firms was also concerning:
- The median Russell 3000 firm recorded a minor remeasurement and repatriation tax charge, amounting to 0.04% of assets or 0.05% of pre-TCHA equity market value.
- Firms in the 10th and 90th percentiles instead recorded a tax benefit or expense of 3.01% and 3.03% of total assets, respectively.
The graph below separates out the two nonrecurring components. Remeasurement charges resulted in large nonrecurring tax expenses in defence, recreation and trading industries (Panel A). The largest average deemed repatriation taxes (Panel B) are found in apparel, candy and soda, and recreation.
Chart 3: Recreational industries suffered most
Five Characteristics That Make a TCJA Winner
Most US firms enjoyed large tax savings, but big winners displayed similar characteristics. Companies that were more likely to experience an effective tax decrease had the following:
- Large initial ETRs – A firm’s past ETR is by far the most important driver of the change in its ETR from before to after the reform.
- Smaller foreign revenues – This is mainly because foreign profits do not directly benefit from the reduced US statutory tax rate.
- Larger capital expenditures – These firms benefited from a reduction in Cash ETR, with no change in GAAP ETR. This comes down to capex expensing deferring taxes.
- Higher R&D expenditures – To some extent this reflects R&D tax credit rules – higher R&D expenses means credit will offset a larger chunk of its pre-credit taxes.
- Size advantage – Larger firms enjoyed larger reductions in tax rates. This is strengthened when foreign revenue is added.
These five characteristics do, however, only explain half of the variations across changes in the ETRs for firms.
Why Did the Market Get It Wrong?
To investigate stock price behaviour, the authors use firms’ cumulative CAPM-adjusted abnormal returns from 2 November – 22 December 2017. They run three specifications: (a) a ‘proxy’ specification, which includes ex-ante variables that may have been used to predict TCJA impacts, (b) an ‘actual’ specification, which includes the ex-post effects mentioned in section 3, and (c) a horse race specification with variables from the previous two combined. The results:
- Prior to the TCJA, markets found the following proxies important to be indicators of expected impacts (as reflected by movements in returns between the dates above): high-tax firms and firms with large capital expenditure were predicted to be beneficiaries, while those with a large share of foreign revenue and large R&D expenditures were expected to lose out.
- After the release of financial statements, the authors found repatriation to be the main driver of returns – firms with large repatriation taxes lost significantly. A 1% of assets change in repatriation tax leads to a 0.8% fall in cumulative abnormal return.[1]
- The same results hold for (a), but the results for (b) become statistically insignificant.
The authors argue that these results suggest that investors priced the impact of the TCJA using proxies[2] rather than more precise estimates. As a result, stock market returns would not have priced in all the relevant information. In theory, the part that the proxies failed to capture should be reflected in subsequent stock price changes as more information became available.
Arguably, markets will have gleaned much information about the TCJA’s impact during the first quarter of 2018. Therefore, stock prices should impound relevant information by the end of the second quarter. Indeed, during Q1 and Q2 of 2018, changes in a firms’ stock return become more correlated with information contained within its financial statements (the proxies become insignificant). Otherwise stated, the market strongly revised downwards the value of firms that would turn out to experience higher recurring taxes.
Bottom Line
The Tax Cuts and Jobs Act of 2017 undeniably decreased effective tax rates across the US. This research, however, leads to the conclusion that the TCJA did not reduce inequality in ETRs. In fact, smaller Russell 3000 firms were more likely to face tax increases, especially if they operated within real estate, recreation or trading. That firms which lost out on effective tax decreases were also more likely to be faced with nonrecurring tax expenses compounded this inequality.
Note that the ‘communications’ sector, which includes FANG companies, was found to be a net loser in terms of the recurring impacts of the TCJA. Effective tax rates are estimated to be approximately 3% higher after the act was introduced. Companies in this sector were, however, less affected by nonrecurring income taxes. Indeed, they were among some of the largest winners, with nonrecurring tax benefits amounting to 1.5% of assets. Relatively smaller repatriation tax charges drove this predominantly.
From an investor perspective, the authors encourage caution. The markets’ inability to readily impound the uncertain consequences of the TCJA could be generalised to other contemporary events, such as pandemics and civic disruptions. Perhaps the most valuable takeaway, however, is what to expect if the next US administration winds down the TCJA. You can find details and analysis of Biden’s tax proposals here.
Reference
1. This result states that firms with larger repatriation costs experienced larger downward adjustments in abnormal returns. It does not say that firms did not benefit from reinvesting repatriated earnings; indeed, Atwood et al. (2020) find firms that faced higher repatriation tax pre-TCJA conducted more acquisitions post-TCJA. Otherwise stated, the results read as such: when controlling for other recurring and non-recurring tax impacts, a firm that needed to pay more tax on foreign earnings experienced lower abnormal returns.
2. Proxies include: foreign revenue, capital expenditures, R&D expenditures, interest deductibility, tax loss carryforwards, repatriation costs and net DTL.
Paper: The Tax Cuts and Jobs Act: Which Firms Won? Which Lost?
Authors: Wagner A. F., Zeckhauser R. J. and Ziegler A.
Publication: NBER working paper no. 27470
Publication date: July 2020
Weblink: https://www.nber.org/papers/w27470
PDF: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=36297
Sam van de Schootbrugge is a macro research economist taking a one year industrial break from his Ph.D. in Economics. He has 2 years of experience working in government and has an MPhil degree in Economic Research from the University of Cambridge. His research expertise are in international finance, macroeconomics and fiscal policy.
(The commentary contained in the above article does not constitute an offer or a solicitation, or a recommendation to implement or liquidate an investment or to carry out any other transaction. It should not be used as a basis for any investment decision or other decision. Any investment decision should be based on appropriate professional advice specific to your needs.)