OBBA polls poorly
Conventional political wisdom says tax cuts are popular. People don’t like paying taxes and do like when politicians cut them.
TCJA was an exception. For example, a 2018 Pew Poll showed 37% approved and 46% disapproved of its overall long-term effect on the country (without the long-term qualifier the gap was slightly smaller).
Unlike TCJA, OBBBA was not just a tax cut, but a budget bill that cut spending (albeit off in the future) to partially offset the lost revenue from the tax cut. That makes it more challenging politically. Not surprisingly, polling typically finds an approval rate lower than TCJA’s. Pew found 33% approval and 46% disapproval, a 3-percentage point larger gap than for Pew’s 2018 TCJA poll. Small difference and, of course, both are strongly correlated with partisan affiliation.
Refunds to the rescue?
This has the GOP worried.1 So, their natural reaction is revise how they are marketing it. Everything is a marketing problem. First step is rebranding (i.e., to change the weird name that the Marketeer in Chief gave to it). The NYTimes reports they have settled on Working Families Tax Cut.2
The second step is to stick your head in the sand: “Let’s assume it’s not really a problem because tax cuts are popular.” In a version of this, the Times story says that Republicans are thinking OBBBA’s big refunds will be their political salvation, since the IRS has not put out new withholding tables. According to the Times:
Republicans are also banking on the simple power of direct payments to ultimately buoy the law’s popularity and, in turn, the party’s prospects in midterm elections next year. Under their design, the law will first deliver many of its benefits to American in their tax refunds next year, a lump-sum payment that may make the tax cut particularly visible to voters.
Most of the “cut” is not new, but just an extension of what people already are used to, i.e., the provisions of TCJA that otherwise would have expired. Put another way, TCJA is obviously reflected in current withholding, and its extension won’t generate refunds. Yes, OBBBA layered on new cuts, but they were relatively modest and targeted to narrow constituencies, as the Times story points out:
Republicans did layer some additional tax cuts on top of the extension. While the legislation overall is unpopular, several of these specific ideas, like tax breaks for overtime pay or tipped income, poll well with both parties. But those cuts will be valuable to only a relatively small subset of Americans, not the population overall, potentially limiting their political resonance.
About 3 percent of American workers regularly earn tips, for example, though even some of those workers may not gain anything from the change. Only Americans who owe a lot in state and local taxes will benefit from an expansion of the state and local tax deduction. And only those who buy a new car made in the United States will be able to deduct their auto loan interest.
Favored groups may notice
I thought I would dig a little deeper. The Yale Budget Lab calculated the differential impact of the new provisions versus the TCJA extension. It reports (emphasis in original):3
- Many taxpayers will see little-to-no additional tax relief. For tax year 2026, we estimate that about one-third of households will see no additional benefit on top of TCJA extension. Almost half will see a tax cut of less than $100 for the year, and two thirds will see a cut of less than $500.
…
- Notable tax cuts are most common in the upper-middle income range. More than half of taxpayers in the fourth income quintile (about $75,000 to $130,000) would see a tax cut of at least $500, and half of top-quintile taxpayers will see a tax cut of at least $1,000. These groups are generally more likely to benefit from the “no tax on…” provisions (including the new senior deduction) and the higher SALT cap.
Tax cuts of $100 will be hardly noticeable. The exact amount of income tax one pays or their typical refund are not very salient, unlike the price of gas which is on big signs everywhere. I would guess most folks only have a general idea of the size of their refund and $75 to $125 more will not particularly stand out. However, a tax cut of $500, which YBL estimates one-third will receive, should be noticed. But I doubt that will matter much politically because of the distribution issue described below.
Because the Joint Tax Committee (JCT) staff prepared estimates under the usual method (present law baseline) and under the Senate’s reconciliation work around (current policy baseline), it easy to filter out the estimated differences in the revenue loss for the TCJA extension versus the new tax cuts.
The new tax cuts for individuals look small compared with the total revenue loss from both the extension and the new cuts. To crudely focus on the tax year 2025 effect, I compared the FY 2025 and FY 2026 numbers.4 The total estimated revenue loss was just under $600 billion. By contrast, the new income tax cut for individuals, excluding the business provisions, was about $170 billion, less than a third of the cost.
Probably the bigger problem for the GOP banking on the political benefits of refunds is the concentration of the benefits of OBBBA’s new tax cuts.
The table below lists the provisions, ranked by JCT’s estimates of the revenue loss (a proxy for the tax cut amount). The dollar amounts are for federal fiscal years 2025 and 2026 – i.e., for all of calendar year 2025 and the first three quarters of calendar year 2026. This will capture some of the tax year 2026 effects, since the JCT estimates reflect that people will adjust their estimated payments and withholding during January through September of 2026.
| Provision | $ amount | % of total |
| Expand SALT deduction | (39,247) | 22.8% |
| No tax on overtime | (32,806) | 19.1% |
| Senior deduction | (32,314) | 18.8% |
| Increased std deduct | (26,503) | 15.4% |
| No tax on tips | (10,121) | 5.9% |
| Enhancement of child credit | (10,014) | 5.8% |
| Car loan interest deduction | (7,332) | 4.3% |
| Trump accounts | (7,177) | 4.2% |
| Enhanced rates | (4,948) | 2.9% |
| Enhanced adoption credit | (608) | 0.4% |
| Enhanced child and dependent tax credit | (409) | 0.2% |
| Enhanced dependent care assist | (365) | 0.2% |
| Enhanced employer-provided child credit | (45) | 0.0% |
| Total | (171,889) |
GOP political payoff?
The provisions with the largest costs confer benefits on narrow categories of taxpayers. The standard deduction and rate changes apply broadly but are less than 20% of the cut. The senior deduction (another 19%) applies somewhat broadly (TPC estimates about 13% of tax units and half of those over 65). The rest of the cuts – over 60% – are narrowly targeted. Proportionately few households earn material overtime, receive tip income (and have tax), take out a car loan, or had a baby during the year (to get a $1k Trump account).5
Expanded SALT deduction
The largest benefit, the increase in the SALT deduction limit to $40k, will narrowly benefit mainly upper middle-income households and often in blue states. Here’s the Yale Budget Lab graph.

Tips and Overtime
The politics are probably going to come down to the refunds generated by the exemptions for tip and overtime pay income, two of Trump’s big headline political points during the campaign.
A preliminary TPC estimate is that the exemption for tip income will benefit 3% of tax units and for overtime pay about 9%.6 See the TPC graph below for the distribution of the benefit by income:

This TPC graph shows the distribution of tip income workers by state.

Overtime pay provides a bigger buck exemption and is likely to generate bigger refunds but is also harder to get a handle using government data on who will benefit. The Yale Budget Lab estimates that about 60% of employees qualify for overtime under the Fair Labor Standards Act, but only 8% of hourly employees and 4% of salaried employees regularly collect it.
Midterm benefits?
The political question is whether biggish refunds from expanded SALT deductibility, tips, and overtime pay will help overcome the general antipathy to OBBBA. The crucial focus must be on persuadable or swing voters in the polarized political world we now live in. The vast majority (maybe 90%) of voters are going to vote based on the candidate’s party in almost all cases. That’s even more likely in low-turnout midterm elections. Moreover, these illusive swing voters only matter in the handful of swing districts for House races and states with Senate seats that are actually in play.
Some relevant factors that occur to me:
- The benefits (refunds?) from the expanded SALT deduction will be concentrated at the top. Those are not the voters that the Republicans have been gaining ground with; instead, they have been migrating to the Dems. What’s worse for the GOP, the beneficiaries will be disproportionately in blue and purple states. That might help them in arguable competitive Senate races in blue or purple states (Maine, Minnesota, New Hampshire, and North Carolina).
- Per the TPC graph, tipped workers are concentrated in red states with Nevada and Wisconsin, two purple states, being the obvious exceptions. (Hawaii is a deep blue state that also has a lot of tipped workers. It’s irrelevant to the political calculations.) That is obviously not a good thing politically for the GOP if they’re looking for an advantage in the midterms. No Senate seat is up in either Nevada or Wisconsin.
- Tipped workers are heavily female. According to TPC, over 70% of tipped workers are female. Of course, the Dems do much better with female voters and especially women of color (over 29% of tipped workers according to TPC), the most reliable of the Dems’ base voters. Will the exemption cause a material number of them to instead vote for Republicans? I doubt it. That is probably not good news for the GOP.
- Gaming out the effect of the overtime pay exemption is more difficult because of the lack of data. My perception is that unionized workers in the private sector, particularly in manufacturing, and public safety employees (cops and firefighters) are the prime recipients of regular and larger amounts of overtime pay. The GOP already does well with both groups. Cops and firefighters are the outliers among public employees – their partisan allegiance tends Republican. It doesn’t seem like a fertile hunting ground to look for voters to swing from the Dems to the GOP.
- More fundamentally, I’m dubious about the political benefits of modest increases in income tax refunds. I suspect that most voters look to the future, rather than rewarding politicians for what they have already done. The tips and overtime pay were modestly big issues in 2024 campaign, Trump won, and the expectations are already baked in. I don’t think the Dems got much mileage in the 2022 midterms out of their generous increase in the child tax credit in the 2021 American Rescue Plan. That expansion had a much broader and more dramatic effect. Its benefits ($85 billion/year) exceeded in dollar terms the sum of the exemptions for seniors, tips, and overtime pay.
Bottom line: I don’t think refunds will bail out GOP from adverse public perceptions of OBBBA. That does not mean opposing OBBBA will be a winning midterm campaign issue for Dems, though. That will depend upon political messaging, media coverage, campaigning, and (potentially) random events (recall how 9/11 threw the 2002 midterms to the GOP).
I know little about effective political messaging, but I’m highly skeptical of what appears to be the Angie Craig approach (see note 1) that muddies what should be unambiguous Dem opposition to OBBBA. Key points in my amateur attempt:
- Unaffordable: We can’t afford to borrow money to hand out more tax cuts. That’s what OBBBA does. Federal debt is at unsustainable levels. To the extent they’re paying with tariffs (plenty Trump quotes to cite), that’s anti-growth and a tax that falls heavily on low- and middle-income people.
- Unfair: OBBBA is both tilted to the top and picks favorites (tips, overtime pay, car loans, etc.). Sure, tipped employees are deserving and often underpaid, but why should they pay less tax than a factory or warehouse worker with the exact same income? Unless you believe in the fiscal fairy, exempting some means everybody else pays more or gets less.
- Cuts crucial services: Cuts to health care (Medicaid and failure to extend ACA tax credits) and food aid (SNAP) hurt many people across all parts of the country, rural and urban, Republican and Democratic.
- In short, don’t do anything to imply OBBBA may be okay/good, if only they had modestly changed its emphasis.
Silver Lining
One glimmer of light is that the administration’s political concerns about making sure that OBBBA is administered (i.e., the 2025 tax filing system works smoothly and the refunds get paid) likely means that the IRS will be saved from shutdown revenge reeked on “Democrat agencies.”
That appears to be the case so far. See this table from the NY Times that shows only 2% of Treasury employees have been furloughed:
| Agency | Total employees | Planned furloughs | Share |
|---|---|---|---|
Environmental Protection Agency | 15,166 | 13,432 | 89% |
Education | 2,447 | 2,117 | 87 |
Commerce | 42,984 | 34,711 | 81 |
Labor | 12,916 | 9,792 | 76 |
Housing and Urban Development | 6,105 | 4,359 | 71 |
State | 26,995 | 16,651 | 62 |
Energy | 13,812 | 8,105 | 59 |
Interior | 58,619 | 30,996 | 53 |
Agriculture | 85,907 | 42,256 | 49 |
Defense (civilian work force) | 741,477 | 334,904 | 45 |
Health and Human Services | 79,717 | 32,460 | 41 |
Small Business Administration | 6,201 | 1,456 | 23 |
Transportation | 53,717 | 12,213 | 23 |
Social Security Administration | 51,825 | 6,197 | 12 |
Justice | 115,131 | 12,840 | 11 |
Office of Personnel Management | 2,007 | 210 | 10 |
Homeland Security | 271,927 | 14,184 | 5 |
Veterans Affairs | 461,499 | 14,874 | 3 |
Treasury | 81,165 | 1,736 | 2 |
This Bloomberg article says that if the shutdown persists, IRS plans to lay off 35,000 employees but not those “working on filing season activities, implementing legislation, and IT modernization.” Getting refunds out and implementing their new law is “essential” collecting tax (auditing and exam) is not. Got it.
Notes
- Perversely, some Dems, including Angie Craig, are worried (WSJ) but for the opposite reason – parts of OBBBA poll well (e.g., exempting tips and breaks for seniors). The response is to expand them. Yikes. Bad policy idea and it endorses a fundamental flaw in the OBBBA’s strategy, government borrowing to hand out tax cut goodies to favored constituencies. It’s bad both because we can’t afford them and they’re unfair, treating people with the same incomes unequally. All this reflects the uncertainty of political winds and why politicians that try to sail with them, rather than expressing their core convictions (e.g., like AOC and Bernie on the left or Liz Cheney on the right do), look like Caspar Milquetoast. Not what inspires confidence in a leader IMO. ↩︎
- I’m sure that the Dems will be in the rebranding business too, claiming OBBBA really stands for One Big Billionaires’ Bailout (or Benefit) Act or something catchier. ↩︎
- Because these estimates are for next year (2026), they are likely slightly higher than the 2025 refund effect, but close enough for our purposes. ↩︎
- That captures a fair amount of the tax year 2026 effect as well, unfortunately. ↩︎
- That assumes they know enough to apply for a Trump account for their baby. Will applications be included in hospital packages for new parents? I would not be surprised if the administration provides a presidential letter claiming responsibility – similar to Trump’s rebate letters during COVID. ↩︎
- Of course, it would be no surprise if a lot more tip income miraculously appears (compared to the estimates), once it is exempt. Congress left many issues to the IRS and there is a fair amount of grey area for the aggressive to game, legally or illegally. ↩︎
Environmental Protection Agency
Education
Commerce
Labor
Housing and Urban Development
State
Energy
Interior
Agriculture
Defense (civilian work force)
Health and Human Services
Small Business Administration
Transportation
Social Security Administration
Justice
Office of Personnel Management
Homeland Security
Veterans Affairs
Treasury

