US Daily: More Spending, More Growth (Phillips/Briggs/Mericle)
Table of Contents
The American Rescue Plan (ARP) totaled $1.84 trillion (8.4% of GDP), somewhat larger than the $1.5 trillion (6.8%) we have been assuming in our economic forecast. Stimulus payments and child tax credits account for $120bn of the gap. The rest of the difference reflects items that should have less of a near-term economic impact, such as fiscal transfers to state and local governments, transit agencies, and pension funds.
We have made some changes to our longer-term fiscal assumptions in light of the new law. Specifically, we now assume that Congress will extend the larger child tax credit past its year-end expiration and that expanded unemployment insurance eligibility and benefit duration will last through 2022.
Details on the next fiscal package are scarce. At this point we expect the White House to propose at least $2 trillion for infrastructure, though this could reach as much as $4 trillion if the proposal extends to other areas (e.g., child care, health care, or education). Increases in the corporate and capital gains tax rates look likely to finance a portion of this, though we believe it will be difficult for Congress to agree on more than around $1 trillion in such offsets.
While the next major fiscal proposal might come with a large headline number, it is likely to have a much smaller impact on growth in 2021 and 2022, as the spending will be more evenly distributed over several years and some of it will likely reflect spending that would have happened in any case. Offsetting tax increases could also incrementally dampen the effects.
In light of the larger fiscal package just enacted, we now expect slightly higher GDP growth of +6% / +11% / +8.5% / +6.5% in 2021Q1-Q4 which implies +7.0% in 2021 on a full-year basis (vs. +5.5% consensus) and +8.0% on a Q4/Q4 basis (vs. +6.0% consensus).
As we now expect a somewhat slower drop-off in fiscal support in subsequent years, we have also raised our GDP growth forecast in 2022 by 0.6pp to +5.1% on a full-year basis (vs. +3.8% consensus) and by 0.5pp to +2.9% on a Q4/Q4 basis (vs. +2.6% consensus).
We have also nudged down our unemployment rate forecast and nudged up our core PCE forecast. After these changes, we view the first hike in the funds rate as a close call between the second half of 2023 and the first half of 2024, and the timing will depend critically on where FOMC participants at that time put the core PCE liftoff threshold. The summary of economic projections provided at the March FOMC meeting next week will provide insight into the FOMC’s reaction function, and we will update our Fed call in our March FOMC recap.
More Spending, More Growth
President Biden signed the American Rescue Plan (ARP) on March 11, just ahead of the expiration of unemployment benefits on March 13. Now that the bill has been enacted into law, we are updating our fiscal assumptions and our economic forecasts.
The final bill was closer to the original Biden proposal than we expected. Our forecast has included a $1.5 trillion (6.8% of GDP) package, or around $350bn (1.6% of GDP) less than what Congress ultimately approved. Exhibit 1 compares our prior expectations and the final package as estimated. Greater-than-expected fiscal support for state and local governments, transit systems and pension funds account for around half of the difference. As discussed below, we expect these provisions to have a modest growth effect.
Exhibit 1: The Size of the Final Phase 5 Bill Was Close to the Original Proposal
Source: Congressional Budget Office, Joint Committee on Taxation, Treasury, Goldman Sachs Global Investment Research
Provisions affecting households account for most of the remaining difference. Most importantly, the final package includes around $120bn more in stimulus payments and child tax credit expansion than we assumed. This is likely to make more of a difference later this year and in early 2022, as Treasury is expected to begin making monthly payments equal to half the value of the credit starting in July; taxpayers will claim the other half on their tax return in early 2022.
The remaining differences in household-sector provisions are likely to be less important to the near-term consumption outlook. The CBO estimate of the bill shows roughly $50bn more in benefits than what we have assumed, but this relates almost entirely to the higher unemployment rate that CBO projects rather than a policy difference. The remaining difference of around $100bn relates mainly to rental/homeowner assistance and health insurance subsidies.
The budgetary savings provisions were also somewhat greater than expected and might provide some clues about where Congress might look for savings in upcoming fiscal legislation. The bill includes around $70bn in tax increases aimed mainly at multinational corporate income, executive compensation, and a limitation on business losses that partnerships and other non-corporate taxpayers may deduct. It also increases the rebates pharmaceutical companies must pay the Medicaid program. These provisions were added in order to offset last-minute changes to the bill as well as to offset the increase in the deficit the bill would have caused after 2031, which Senate rules require for any provisions in reconciliation legislation. Congress is likely to revisit some of these areas when seeking offsetting savings for the next fiscal package.
The fiscal effects are very front-loaded, but less than the original version of the bill or the headline numbers imply. Based on the final Congressional Budget Office (CBO) estimate with a rough translation to calendar years, the bill looks likely to increase the deficit by roughly $1.25 trillion in CY2021, $475bn in CY2022, and around $125bn in CY2023. However, the economic impact is likely to be less front-loaded than this, as some of the funds that the federal government will “spend” this year reflect fiscal transfers to other entities that will take much longer to spend the money. For example, the bill provides $350bn of fiscal aid to state and local governments, half of which the Treasury will grant in the next 60 days. In light of the better-than-expected state budget situation and substantial fiscal aid Congress already approved, states are likely to take several years to spend these funds, even though they will show up in the federal deficit almost immediately.
We expect this to be the last major COVID-focused fiscal relief package, but Congress seems likely to renew some of the items in it later this year. Congress looks likely to extend two items, in particular. First, the expansion of the child tax credit in this bill is likely to prove popular and we expect congressional Democrats to press to extend it or to make it permanent. We have built in an extra $110bn in annual child credit payments into our forecast to account for this. Congress also looks likely to further extend some expanded unemployment benefits. In addition to the extra $300/week payment ARP provides, the bill also renews expanded eligibility and further extends the maximum duration of benefits, to 79 weeks. We assume that Congress will allow the $300/week benefit to expire in September, but we expect the enhanced eligibility and benefit duration policies to last through 2022.
The next round of fiscal legislation could have a large headline cost but it will probably have a much smaller near-term effect than the bill Congress just passed. As we recently wrote, we expect the next round of infrastructure legislation to increase spending and tax incentives for traditional and green infrastructure by at least $2 trillion. If the Biden Administration expands the bill to include other policies like expanded child care benefits, extension of the child tax credit, student debt relief or other education subsidies, and/or health insurance subsidies, the gross amount could rise to something like $4 trillion.
However, these large numbers are not comparable to the bill Congress just passed, for three reasons. First, most of these policies would cover several years and some could even be permanent. While the ARP costs $1.84 trillion over ten years, most of the budgetary impact will be in the first two years. By contrast, a $2 trillion (over 10 years) infrastructure package would probably raise spending by less than $200bn in the first year after enactment. That said, if lawmakers add policies in other areas to the bill, like an extension of the newly-expanded child tax credit, the near-term fiscal impact could be greater.
Second, the headline price of an infrastructure package is likely to count some ongoing spending. For example, last year’s $1.5 trillion House-passed infrastructure bill counted nearly $500bn of highway/transit spending toward the total, though around $300bn of that would have been spent regardless of the legislation. A smaller but similar situation exists with green tax incentives.
Third, some of the spending is likely to be offset by tax increases and other spending cuts. We expect that an increase in the corporate tax rate (perhaps to 24-25%), an increase in the capital gains rate (perhaps to 28%) and various other tax provisions will offset a portion of the cost of the next fiscal package. A smaller contribution might also come from cuts to health spending, particularly items related to drug pricing. We doubt that Congress will be able to agree to more than around $1 trillion (over ten years) of tax increases and other budgetary savings.
We expect an outline of the Administration’s plans in a few weeks. President Biden is likely to address a joint session of Congress in April, at which point we expect he will outline plans for infrastructure and the remaining parts of the Administration’s agenda. Around this time, and certainly by May, we expect the White House to submit a budget proposal to Congress, which should provide even more detail.
Legislation looks likely to emerge in May or June. Following the rollout of the Administration’s plans, we expect three steps. First, Democratic committee chairs and other relevant Administration officials are likely to negotiate with congressional Republicans on a potential agreement. This is a necessary political step, we believe, in light of the President’s stated desire to enact bipartisan legislation and the likely resistance among centrist Democrats to immediately pursuing legislation via the reconciliation process (which requires only 51 votes in the Senate). We are skeptical that the entire infrastructure package can pass with bipartisan support, as Republicans seem likely to object to the size of such a package as well as rolling back parts of the 2017 tax law to pay for it. Assuming that Democrats ultimately decide to rely on only Democratic votes by using the reconciliation process, we would expect the first legislative step—passage of the FY2022 budget resolution that instructs the relevant committees to pass legislation that achieves specific fiscal targets—to occur in May, though the timing will depend on how long bipartisan discussions last.
It is possible that Democratic leaders could split off some infrastructure elements and pass them with bipartisan support. While we expect most of the infrastructure package to pass via the reconciliation process, some parts might move separately. For example, the 5-year reauthorization of the federal highway/transit program (the “highway bill”) expires in September, and it would be nearly impossible for congressional Democrats to pass via reconciliation due to Senate rules. This is also an area where bipartisan cooperation has occurred in the past and seems possible again this year.
September will be the deadline for the infrastructure bill, though it could pass earlier. As noted above, the highway bill expires Sept. 30, which is likely to become a working deadline for the broader infrastructure package. If the legislative process starts in May and there are no major delays, it is conceivable that Congress could enact the package in late July or early August, ahead of the August recess. However, the more likely scenario in our view is that the bill is enacted around September.
Our revised fiscal assumptions imply a larger fiscal impulse in 2021 and especially 2022. To evaluate the growth implications of the fiscal upgrade we have updated our estimates of the impact of fiscal support on the level of GDP. These estimates, shown in Exhibit 2, incorporate the effects of some residual CARES Act measures that will be spent in 2021, the Phase 4 package, the Phase 5 package, and the UI benefit extensions, child tax credit extensions, and infrastructure spending that we expect to be passed later this year. We expect that the impact of fiscal stimulus on the level of GDP will peak between 6-7% in 2021Q2 when the $1400 stimulus checks have their largest effect before declining for the remainder of our forecast horizon. Relative to our prior fiscal assumptions, we now expect a much slower drop-off of support past 2023Q3 and especially in 2022 when the increased spending in the Phase 5 fiscal package has its largest impact and the spending we expect will be passed later this year begins to be disbursed.
Exhibit 2: The $1.84tn Package Implies a Larger Fiscal Impulse in 2021 and Especially in 2022
Source: Congressional Budget Office, Joint Committee on Taxation, Department of Commerce, Treasury, Goldman Sachs Global Investment Research
We have also incorporated the upgraded fiscal outlook in our growth forecast. We now expect GDP growth of +6% / +11% / +8.5% / +6.5% in 2021Q1-Q4 (vs. +5.5% / +11% / +8.5% / +6.0% previously), implying 2021 growth of +7% on a full-year basis (vs. +6.9% previously and +5.5% consensus) and +8.0% on a Q4/Q4 basis (vs. +7.7% previously and +6.0% consensus). We now expect a smaller sequential deceleration in 2022, and have raised GDP growth to +5.1% on a full-year basis (vs. +4.5% previously and +3.8% consensus) and +2.9% on a Q4/Q4 basis (vs. +2.4% previously and +2.6% consensus). That said, the positive growth effects are concentrated in the first half of 2022, and we continue to expect growth in 2022 H2 to slow to 1.75%.
Exhibit 3: We Still Expect Well Above-Consensus Growth in 2021, but Now Expect Less of a Slowdown in 2022
Source: Bloomberg, Goldman Sachs Global Investment Research
We have also nudged down our unemployment rate forecast—which was already the lowest among forecasters included in the Bloomberg consensus survey at the end of 2021—to reflect our more positive growth outlook. We now expect the unemployment rate to decline to 4.0% / 3.5% / 3.2 %/3.1% at the end of 2021-2024 (vs. 4.1% / 3.7% / 3.4% / 3.2% previously), thereby reaching last cycle's 50-year low at the end of 2022 and falling further in subsequent years.
We have similarly nudged up our inflation forecast to reflect the tighter labor market, and now expect core PCE to reach +1.90% / +2.00% / +2.1% / +2.15% at the end of 2021-2024 (vs. +2.00% / +1.90% / +2.05% / +2.15% previously). This change reflects a tighter labor market, as well as an earlier-than-expected end to crisis-related Medicare payments (at the end of this month, vs. our previous expectation that the Phase 5 bill would be renew these payments for the rest of 2021).
After these forecast upgrades, we view the first hike in the funds rate as a close call between the second half of 2023 and the first half of 2024. Our working assumption has been that the Fed’s implicit core PCE inflation threshold for liftoff is 2.1%, and we now expect to reach this slightly earlier in 2023H2, which would argue for bringing forward the timing of liftoff from our standing forecast of 2024H1. However, we are increasingly unsure where FOMC participants put this threshold and we see the risks as tilted to the higher side. The summary of economic projections provided at the March FOMC meeting next week will provide insight into the FOMC’s reaction function, and we will update our Fed call in our March FOMC recap.
Alec Phillips
Joseph Briggs
David Mericle
Investors should consider this report as only a single factor in making their investment decision. For Reg AC certification and other important disclosures, see the Disclosure Appendix, or go to www.gs.com/research/hedge.html.