“Here’s What’s In It”: Goldman Explains All You Need To Know About The Current State Of Tax Reform
To the delight of Donald Trump, just before 2am on Saturday morning the Senate passed the Republican Tax bill in a 51-49 vote, and with tax reform legislation now passing both chambers of Congress it looks very likely to become law by year-end, probably within the next two weeks according to Goldman Sachs which now ascribes a 90% probability of the legislation becomes law by year end.
In it latest assessment of the state of tax reform, Goldman analysts Alex Phillips and Jan Hatzius write that while largely a done deal, some differences between the two versions still need to be ironed out: “We expect the final structure of the bill to reflect more of the Senate bill than the House bill, including a 20% corporate tax rate effective in 2019, the Senate’s more restrictive limit on net interest deductibility, and the Senate’s treatment of pass-through income. Both proposals now include a $10k cap on state and local property tax deductibility, rather than full repeal, eliminating the most important political difference between the bills before the conference negotiations start.”
Additionally, Goldman adds that while the corporate tax changes are likely to result in a net tax reduction in corporate tax liabilities, the size of the tax cut actually looks fairly small. Compared to current policy, the compromise legislation we expect to emerge from the conference committee would reduce the effective corporate tax rate by only a couple of percentage points.
Perhaps the most surprising take home from the Goldman analysis is that while the bank has increased its estimate of the growth impact from tax reform slightly, to around 0.3% in 2018 and 2019 – “reflecting the slightly larger amount of tax cuts in the Senate plan following revisions, and our expectations regarding the eventual compromise” – it still expects a relatively modest boost to overall economic growth.
That said, questions remains, most notably: “what’s in the actual bill?”
To answer, we publish the latest Goldman analysis for those still confused – which would be pretty much everyone – what is currently contained in the most sweeping tax overhaul in the US since the days of Ronald Reagan.
Tax Reform: The Home Stretch
Q: The Senate has passed the bill, now what?
A: Differences between the House and Senate bills are likely to be reconciled in a conference committee. A conference committee involves the appointment of conferees of both parties from the House and the Senate, with a majority of conferees needed to approve the final agreement. As a practical matter, House and Senate Republican leaders and a few other relevant Republican lawmakers are likely to negotiate the final agreement, as recent votes in the House and Senate demonstrate that Democratic support is unlikely to be needed to conclude the conference negotiation. Once the final conference report has been filed, the House and Senate must pass it once again before sending it to the President for signature. A simple majority would be required in both chambers, with no changes possible.
A possible alternative would be for the House to simply pass the Senate-passed bill, avoiding the conference process and expediting enactment. In light of the impending special Senate election in Alabama, upcoming fiscal deadlines, and general political uncertainty, congressional Republican leaders might consider this option if conference negotiations take longer than expected, though at this stage a conference committee looks much more likely.
We expect congressional Republican leaders to begin conference negotiations immediately, and believe they will target completion the week of December 11. If successful, this would produce final details around December 11-13, and final passage in the House and Senate December 14-15. One reason we expect this timing is because of the need to address expiring spending authority by December 8, which we expect to be extended temporarily through December 22, creating only a short period before year end when Congress is not addressing other fiscal deadlines.
Q: How likely is this to become law?
A: It is extremely likely that tax reform legislation becomes law, with a 90% chance it becomes law by year-end. Our view has been that once legislation has cleared the Senate, the odds of enactment would be quite high because the Senate has always represented the greatest obstacle to enactment. Reconciling differences in the conference committee represents a risk, but we do not believe congressional Republicans would allow tax reform to fail after having passed similar versions in both chambers. Even in the event that the conference negotiation bogs down, we expect that the House would simply adopt the Senate-passed version if there were no other alternative, though a compromise through a conference committee looks much more likely at this point.
Although the legislative process has been slower than we expected for the most of the year, over the last couple of months we have been consistently surprised at how quickly congressional Republicans have made progress on tax reform. The final step in the process, the House-Senate conference committee, often takes several weeks to complete; in 1986 it took conferees two months to reconcile differences between House and Senate versions of tax reform legislation, for example. However, it would not be unprecedented for a conference committee on major tax legislation to be completed in less time; the conference process for the 1981, 2001, and 2003 tax cuts took a week or less, for example. With the apparent motivation that congressional Republicans have to finish work on tax reform this year, we expect that a conference agreement between the House and Senate could be voted upon by mid-December. While it is possible that consideration of tax reform could spill over into January, at this point enactment in December looks far more likely.
Q: How does this compare to consensus?
A: Market pricing also reflects a view that tax reform is likely to become law. Over the last few weeks, high-tax stocks have outperformed low-tax stocks (Exhibit 2). This reflects, in our view, a growing expectation of tax reform, which should benefit companies with high effective tax rates more than companies with low effective tax rates. Prediction markets, which as recently as October ascribed only a 20% probability to tax reform being enacted this year, now imply a nearly 80% probability. The implied probability by the end of Q1 is around 95%. Our conversations with clients also suggest little remaining uncertainty regarding whether the bill becomes law. Instead, the focus has shifted to what changes might still be made, how differences between the House and Senate versions will be resolved, and what the effect will be across sectors.
Q: What changed in the Senate bill?
A: The pass-through, state and local tax (SALT), and capex expensing provisions became more generous, while alternative minimum tax (AMT) changes and profit repatriation rates became less generous. Among the major changes the Senate made prior to committee-passed version of the bill:
- Some property taxes would be deductible. The first $10k in state and local property taxes could be deducted under the Senate bill, bringing it into line with the House version. Previously no state and local taxes could be deducted from non-business income under the Senate version.
- The deduction for pass-through income has increased to 23%. The benefit would phase out for taxpayers with income above $500k, similar to the prior version. For taxpayers with total income near the limit who would otherwise be in the 35% bracket under this proposal, this would work out to tax rate on pass-through income of roughly 27%, rather than 29% under the prior version.
- Capex benefits would last slightly longer. Under the prior proposal, full expensing of equipment investment would have expired after 2022. With the recent revisions, the share of equipment that could be deducted in the year of investment would decline by 20pp after 2022, expiring fully in 2027.
- No AMT repeal, after all. The AMT imposes additional tax beyond the standard income tax on middle- and upper-income taxpayers with substantial deductions, among other circumstances. A separate AMT is applied to corporations. The House bill and original Senate bill would have eliminated the AMT; the revised Senate version increases the exemption amount for individuals but stops short of repeal; the corporate AMT appears to be left in place as well. Individual and corporate AMT repeal were estimated to reduce revenues by $770bn and $40bn respectively; this change is expected to offset the cost of some of the more generous provisions noted above.
- Higher tax rates on unrepatriated profits. The prior Senate n proposal would have taxed untaxed foreign profits at 10% if held in cash or liquid assets, or 5% if not. The current version steps up those tax rates to 14.5% and 7.5%.
Q: What happened to the “trigger” idea?
A: The trigger was dropped because it became politically unnecessary. Senator Corker (R-TN) and several other senators who were concerned about the deficit impact of the legislation had proposed a provision that would reverse some of the tax cuts several years from now if revenues had not grown more quickly than the official projections. As the revenue gain from the trigger would have been contingent on economic developments, the trigger was ruled noncompliant with the “Byrd Rule”, which stipulates that only provisions that have a fiscal effect can be included in budget reconciliation legislation. After this ruling, and other changes to the bill, all but one Republican senator had announced public support for the bill, meaning that it had sufficient support without the trigger. It is possible that the concept could be revisited if there is insufficient support for the final conference agreement without it, but at this point a trigger looks very unlikely.
Q: What are the remaining issues that need to be worked out?
A: The greatest policy differences between the House and Senate bills involve the AMT and top marginal rate, pass-through treatment, corporate provisions dealing with cross-border transactions, and net interest deductibility. Exhibit 3 summarizes the differences between the House-passed and Senate-passed versions, along with the estimated revenue effects over ten years estimated by the Joint Committee on Taxation (JCT). The right column of Exhibit 3 suggests what a potential compromise between the two versions might look like that would stay under the $1.5 trillion overall limit on revenue loss imposed by the recently passed budget resolution.
Q: Will the ACA mandate be repealed?
A: We expect the penalty on the uninsured to be set to $0, which would have the same practical effect as repeal. The Senate legislation sets the penalty on the uninsured to $0, which is estimated to generate over $300bn in budgetary savings over the next ten years. This is used to expand other tax cuts in the bill. House Republicans have been more supportive of repealing the individual mandate than Senate Republicans have, so Senate passage suggests that this change is likely to be included in the final version of the legislation, in our view.
Repealing the mandate would have two main effects. First, insurance coverage would decline. CBO has estimated that the level of uninsured would rise by 4 million in the first year after mandate repeal, and by 12 million in the third year.1 Given that the enrollment period for 2018 concludes in less than two weeks on December 15, around the same time that we expect tax reform legislation to become law, our expectation would be that the decline in coverage in 2018 would be somewhat smaller than the CBO estimate but that the effect in later years would be similar.
Second, premiums in the individual market would increase. CBO has estimated that average premiums would rise by about 10% without the mandate. Younger and healthier individuals are the most likely to drop coverage without the mandate. This would leave the remaining risk pool older and less healthy, leading to an increase in premiums. That said, since ACA subsidies are designed so that the government pays the portion of premium that exceeds a certain percentage of an individual’s income, subsidized enrollees would be responsible for only some or, in some cases, none of the additional premium cost. By contrast, unsubsidized enrollees would bear the full increase. CBO has estimated that the individual insurance market would continue to be stable without the individual mandate.
Q: How will corporate interest deductibility change?
A: The outlook here is murkier than in most other areas, but fiscal constraints could lead lawmakers to include the more restrictive Senate proposal in the final version. The House-passed legislation restricts net interest deductibility to 30% of an income definition that roughly translates to earnings before interest, taxes, depreciation amortization (EBITDA). By contrast, the Senate restricts interest deductibility to 30% of an income definition that roughly translates to earnings before interest and taxes (EBIT). The difference is substantial, and JCT estimates that the more restrictive Senate version would generate nearly twice as much revenue as the House provision.
Exhibit 4 shows the average interest deduction by industry as a share of each definition, using 2013 data from the IRS. We note that in the House and Senate bills, utilities are excluded from the limitation; the real estate sector is excluded in the House bill as well, and companies in that sector would have the ability to opt out of the limitation in the Senate bill but would lose the benefit of full expensing if they did.
The outcome for this provision is particularly hard to predict but we believe a provision closer to the Senate provision seems more likely to prevail. In light of the need to offset other changes to the bill, we would expect that negotiators will lean toward the version that generates greater savings if they are able to pass the Senate which has tended to be the higher political hurdle for the tax bill in general.
Q: How will the corporate international provisions be settled?
A: We expect the Senate’s “inbound” provisions to prevail, but the “outbound” provisions are hard to predict. The House and Senate both include “outbound” provisions intended to impose a minimum tax on some of the foreign operations of US companies, and “inbound” provisions intended to combat the erosion of the domestic corporate tax base through transactions with foreign affiliates. The general structure of the outbound and inbound proposals is similar in the House and Senate proposals.
The inbound proposals are conceptually similar but differ in the details. In both cases, they would effectively tax deductible payments that a US company makes to its foreign affiliates. In the House, this is structured as a 20% excise tax, though companies would have the option to elect to be taxed on the associated foreign income instead. In the Senate, the proposal would effectively impose a 10% tax on related-party payments. An important difference is that the Senate provision would appear to exclude payments to related foreign manufacturers for cost of goods sold, while the House proposal could tax some of those payments, with potentially greater effects on cross-border supply chains. That said, even the Senate version is likely to have consequences that are only understood after the legislation has been enacted and companies start to implement the new rules.
The outbound proposals are slightly more straightforward. These would impose a tax on any foreign intangible income exceeding a specified return (e.g., 10% in the Senate bill) on foreign tangible assets (e.g., depreciable assets like equipment and structures). Exhibit 5 shows the effective combined US and foreign tax rate under the House and Senate bills. While the House bill would tax half of this income at the 20% domestic corporate rate, for a 10% effective minimum tax, the Senate uses a more complicated structure that taxes all intangible income from foreign assets as US income at the 20% rate, but provides a 37.5% deduction of all intangible-related income related to foreign sales, whether from US assets or foreign assets. This could remove the incentive to move intellectual property and other intangibles to foreign subsidiaries since they would receive the same treatment on their foreign sales regardless of where the assets were held. However, previous US tax regimes that taxed income from US-based assets differently depending on whether the income was generated by sales in the US or in other markets were repealed after they were challenged successfully in the WTO by trading partners. In light of the risk of another successful challenge, we believe the conference committee is slightly more likely to settle on a policy closer to the House
proposal in this area.
Q: When will the changes take effect?
A: Apart from the potential delay in the corporate rate cut, almost all of the changes will take effect at the start of 2018. There are essentially no retroactive tax cuts or tax increases in the House or Senate proposals, with the notable exception of the tax on accumulated untaxed foreign profits. The only major provision that does not take effect at the start of 2018 is the Senate corporate rate reduction, which remains at 35% in 2018 and drops to 20% starting in 2019. We expect this will be included in the final version, since it reduces the ten-year cost of the bill by more than $100bn. Proponents of the delay argue this would also spur more capital investment in 2018, as it would incentivize companies to pull forward capex that would be fully deductible against the 35% rate in 2018 rather than the 20% rate in 2019. Exhibit 6 shows the overall change in tax receipts estimated by the Joint Committee on Taxation, shifted to a calendar year basis. We note corporate tax receipts are estimated to increase in 2018 under the Senate bill, which results from tax payments related to deemed repatriated profits, which would not be offset by a lower corporate tax rate until 2019.
Q: How will the bill affect corporate tax liabilities?
A: It will reduce effective corporate tax rates much less than the 15pp drop in the statutory rate implies. For context, the JCT estimates of the revenue effects of the tax bills are made against a baseline that assumes roughly $3.9 trillion in corporate tax receipts over the next ten years. This suggests that a corporate tax cut of around $300bn/10yrs should result in only a reduction in the effective tax rate across companies of less than 10%. This would result in less than a 2pp decline in the average effective corporate income tax rate, using for example the 19% average effective rate estimated by the Congressional Budget Office (CBO).
The eventual effect also depends on what assumption one makes regarding the extension of expiring provisions. Under current law, corporate taxes would rise by roughly $250bn over the next ten years due to the expiration of the current 50% bonus depreciation policy for equipment investment and a number of smaller expiring policies would add about $150bn more over the next ten years. Exhibit 7 shows the same estimates of the tax changes shown in Exhibit 6, but adds the effect of expiring policies that have not been addressed in the TCJA.
If Congress took no further action on taxes over the next ten years, corporate taxes would actually increase slightly versus current policy. However, most assume that some of these provisions will be extended when they are set to expire; under the Senate-passed bill, 100% immediate expensing of equipment investment is scheduled to phase down by 20% per year starting in 2023 but Congress could step in to prevent this phase-down.
Q: What does this mean for growth?
A: We expect the legislation to boost growth by around 0.3pp in 2018 and 2019. This is based mainly on the Senate version of the bill, which delays the corporate income tax cut to 2019 but includes a tax cut for individuals (including pass-through income) of around 0.6% of GDP in 2018, slightly greater than what we had previously penciled in. The effect is spread over two years in part because some of the provisions that reduce individual income taxes would show up primarily as lower tax settlements in 2019 rather than reduced withholding from paychecks in 2018.
On the corporate side, we disregard the temporary increase in tax payments in 2018 related to the tax on deemed repatriation; we do not estimate a growth effect from those repatriated profits, either. While the corporate tax cut looks likely to take effect with a one year delay, in 2019, we note that there is likely to be some pull-forward of capex from 2019 into 2018, as companies attempt to maximize their deductions against the higher corporate rate.
We note that the effect in 2020 and beyond looks minimal and could actually be slightly negative, as the JCT estimates suggest that the tax cut that year would actually be slightly smaller than in 2019. Exhibit 8 shows our revised estimate of the growth effects of fiscal policy, incorporating an assumption similar to the fiscal effects of the potential compromise shown in Exhibit 3.
Q: What are the political consequences of tax reform?
A: We do not expect it to help Republican prospects very much. Congressional Republicans have suggested that passing tax reform should help them maintain their majority after the 2018 midterm elections. While passing tax cuts ahead of an election should improve the majority party’s prospects, it is less clear that tax reform will provide the political tailwind Republicans are expecting. Certain provisions are controversial, and the bill overall is relatively unpopular among voters; an average of recent polling shows 32% of voters support the legislation, which compares unfavorably with other major tax cuts like the 1981 or 2001 tax bills, or even the roughly revenue-neutral tax reforms enacted in 1986. This is likely because one in three voters believes their taxes will increase because of tax reform, with more Democrats expecting an increase than average.
If these concerns persist, voters are likely to be ambivalent, or worse, regarding tax reform. That said, sentiment might improve if voters perceive over time that their taxes have declined. The Joint Committee on Taxation (JCT) predicts taxes will decrease or stay the same for about 90% of voters through 2021 (Exhibit 9). The most controversial aspects of the legislation also have mixed implications; while voters favor repealing the individual mandate by two to one, for example, repealing SALT deductions continues to be unpopular and could become a liability in the dozen or so competitive Republican-held House districts in New York, New Jersey and California.