Penn Wharton Budget Model

To evaluate the potential effects of a hypothetical $1.5 trillion Universal Basic Income (UBI) program, PWBM conducts analyses of the program under three different financing policies. Each of the three financing options has different effects on household savings, consumption, and labor decisions, which leads to significantly different effects on the aggregate economy and household welfare: Deficit Financed: In the deficit financed policy, the federal government borrows money to finance the $1.5 trillion in annual UBI transfers. The UBI transfers to households generate a positive income effect, whereby households decide to work less because they are receiving additional income. Furthermore, the higher deficits used to finance this transfer crowd-out productive investment, which leads to lower capital services. Lower capital and lower labor lead to GDP that is 6.1 percent lower in 2027. In this case, lower GDP is reflected in lower total household consumption, which means that the average household is worse off under this policy. Payroll Tax Financed: In this experiment, the federal government levies an additional 11.25% payroll tax on all labor income to finance $1.5 trillion in annual UBI transfers. As in the deficit financed policy, the UBI transfers generate a positive income effect which is reflected in lower household labor, however, this positive effect is partly offset by a negative income effect from higher taxes under the new payroll tax. Since a flat UBI transfer is more progressive than a linear payroll tax, the net income effect comes from shifting some resources from lower marginal propensity to consume households to households with higher marginal propensities to consume. The payroll tax also generates a substitution effect: Lower after-tax wages incentivize households to work less. The substitution effect from the payroll tax is smaller than the net income effect, leading to a smaller decrease in labor compared to the other policies. Furthermore, since the payroll tax covers the cost of the UBI program, there is no direct crowd-out of productive investment. Combined, these effects lead to a smaller (1.7 percent) decline in GDP, and a smaller decrease in household consumption compared to the deficit financed policy. Externally Financed: In this experiment, households receive a transfer of $1.5 trillion annually from an external source, so there is no direct crowd-out of private capital. Like the UBI in the deficit financed policy, these transfers to households generate a positive income effect, but no offsetting negative income effect. There is also no substitution effect. The positive income effect from the UBI transfer, which is larger than the substitution effect from a payroll tax, leads to a larger decline in labor compared to the payroll tax financed policy. Lower labor leads to a smaller economy: GDP is 3.4 percent lower in 2027. However, the externally financed policy has different effects on the components of GDP: consumption, government spending, investment, and net exports. Government spending remains constant. Less wage income reduces investment. Of course, household consumption increases. However, the increase in consumption is effectively offset by a decline in net exports because the UBI is externally financed, thereby acting like a free import. Although GDP declines, households consume more and work less, which improves households’ well-being. The externally financed policy is a good example of why changes in GDP do not necessarily reflect changes in household welfare; even though GDP goes down, the average household is better off because they work less and consume more.

By Richard Prisinzano and John Ricco The passage of the Tax Cuts and Jobs Act brought with it a new 20 percent deduction for income earned from a pass-through business. The IRS recently released proposed regulations that clarify some of the issues associated with the new deduction. Our model suggests that depending on the effectiveness of the regulations, the overall cost of the deduction could be reduced between $54 and $65 billion over the 10-year budget window. Since its passage, there has been considerable discussion about the strategies taxpayers could take to lower their tax liability related to this deduction.1 The legislation defines the income eligible for the deduction as “Qualified Business Income” (QBI). Any income earned from a pass-through business is QBI if the income is earned as an owner rather than as an employee and the taxpayer’s income falls below $157,500/$315,000 (single/married). If a taxpayer’s income is above $157,500/$315,000 (single/married) then the income derived from a Specified Service Trade or Business (SSTB) is not QBI and does not qualify for the 20 percent deduction. Two of the tax strategies that PWBM accounted for in our score of this provision are the conversion of employees to contract workers and ‘cracking’. The first strategy involves employees creating new pass-through entities while maintaining their jobs in order to take advantage of the deduction. The second strategy involves splitting a business into two parts. The first part earns allowable QBI (like rental income or income from custodial services) and charges the highest defendable rate to the part of the business that earns disallowed QBI (like medical or legal services).2 Using this strategy, service providers can move profits away from the SSTB to a business that produces QBI to take advantage of the deduction. The recently released IRS proposed regulations address these strategies. The regulations outline a set of rules to prevent abuse of the deduction. In general, the rules disallow the deduction if the employer-employee relationship pre- and post- conversion does not materially differ.3 In a similar way, the regulations limit the usefulness of the ‘crack and pack’ strategy. The regulations stipulate that if a business earning allowable QBI has at least 50 percent common ownership with an SSTB and provides 80 percent of its “property or services” to the commonly-owned SSTB it earns income from, then the business is considered an SSTB. Additionally, if a business has at least 50 percent common ownership with an SSTB any income earned from that SSTB is ineligible for the deduction. The effectiveness of these proposed regulations will ultimately be decided by IRS action either through audit or litigation. However, it is useful to consider the potential revenue consequences of these regulations. As noted above, our previous estimates account for both of the above strategies. In order to estimate the potential revenue effects we remove certain behavior. We neither allow employees to convert to a pass-through with their current employer nor allow businesses to use the cracking strategy. If the IRS regulations successfully limit this behavior, the former potentially saves $13 billion in revenue over the 10-year budget window while the latter potentially saves $52 billion over the 10-year budget window, each relative to our baseline estimate that includes avoidance behavior. These effects are an upper bound, depending on if the regulations successfully limit behavior, on the revenue gains of the proposed regulations. It is also important to consider the behavior of businesses that can not utilize the deduction. As we noted in our recent brief, those individuals and businesses that can not utilize the deduction have a greater incentive to elect to be taxed as C-corporations. The recent regulations identify those individuals and businesses more clearly. As such, we estimate that $11 billion in tax revenues is lost as these businesses convert to C-corporate status. This behavior yields a potential net revenue gain from the proposed regulations of $54 billion over the 10-year window. --- 1. See Kamin, David, David Gamage, Ari D. Glogower, Rebecca M. Kysar, Darien Shanske, Reuven S. Avi-Yonah, Lily L. Batchelder, et al. “The Games They Will Play: Tax Games, Roadblocks, and Glitches Under the New Legislation.” SSRN Scholarly Paper. Rochester, NY: Social Science Research Network, December 7, 2017. https://papers.ssrn.com/abstract=3084187. ↩ 2. Cracking is often used with ‘packing.’ Kamin et al use the term ‘crack’ to describe the strategy as we present it and ‘packing’ to describe a strategy where a disallowed service goes ‘in-house’ by adding allowable QBI to the SSTB. It is not clear from the regulations whether packing is disallowed. See Kamin et al for more details. ↩ 3. As Tony Nitti notes, the regulations only address current employer-employee relationships; it does not stop employees from starting a relationship with a new firm. See Nitti, Tony. “IRS Provides Guidance On 20% Pass-Through Deduction, But Questions Remain.” Forbes. Accessed August 22, 2018. https://www.forbes.com/sites/anthonynitti/2018/08/09/irs-provides-guidance-on-20-pass-through-deduction-but-questions-remain/. ↩

By Danni Chen In contrast to earlier this year, November has seen consecutive days of falling oil prices. This drop has led to lower costs for consumers at the pump. With the annual growth rate of investment in public infrastructure slowing, some have suggested that now is the time to increase the federal gas tax. Recently, a former Secretary of Transportation urged the Administration to seize the opportunity to raise the gas tax. In the past, President Trump has endorsed an increase of 25 cents per gallon. The last increase in the federal gas tax was a quarter century ago in 1993 to 18.4 cents per gallon. Here, PWBM estimates the amount of additional revenue that is gained by increasing the gas tax by 25 cents per gallon. To estimate the impact of a larger gas tax on revenue, we use the Energy Information Agency’s (EIA) forecast of both retail gasoline price and gasoline consumption. We then adjust gasoline consumption in response to the increased after-tax price of gasoline based on elasticity estimates from the literature.1 The red line in Figure 1 is the forecasted retail gasoline price inclusive of all state and federal taxes, including the proposed 25 cents per gallon increase. Under current law, the federal gas tax is projected to raise an average of $23.7 billion per year from 2018 to 20282 and is represented by the dark blue area. The downward slope of the revenue area reflects forecasted changes to gas consumption due to both changes in the fuel efficiency of automobiles and gas price increases. We forecast the additional 25 cents per gallon tax will raise an additional $31.8 billion per year on average. With the increase, the federal gas tax will generate a total of $55.5 billion per year in average revenue, or $549 billion over the standard 10-year budget window. Figure 1: Estimates of Federal Gas Tax Revenue Under Current Law and with New Tax and Gas Price with New Tax DOWNLOAD DATA --- 1. We use estimated elasticities from Coyle, David, Jason DeBacker, and Richard Prisinzano. “Estimating the Supply and Demand of Gasoline Using Tax Data.” Energy Economics 34, no. 1 (January 1, 2012): 195–200. In applying these elasticities, we assume that retailers pass the entire tax to consumers, which is a standard assumption consistent with the intense price competition in this market. ↩ 2. CBO forecasts a slightly larger average of $24.2 billion per year over the same time period. ↩

By Alexander Arnon Hundreds of thousands of federal government workers were recently on furlough due to a partial lapse in appropriations. Furloughed employees are barred from working and are not paid, but recent legislation guarantees that they will receive back pay “on the earliest date possible after the lapse ends.” The ambiguous status of federal workers on furlough--employed, but not working--will be reflected in next Friday’s jobs report. Furloughed federal workers will be counted as both employed and unemployed in January. How Can Furloughed Federal Workers Be Both? The Bureau of Labor Statistics (BLS) produces the monthly jobs report using two different surveys of employment, known as the establishment survey and the household survey: - The establishment survey asks nonfarm employers to report the number of workers on their payroll--that is, how many workers received wages or salaries for a particular pay period. The change in the number of nonfarm payroll jobs is often referred to as the “monthly jobs number.” - The household survey asks individuals to report whether they or others in their household were working. This second measure of employment is broader than nonfarm payrolls (for instance, it includes the self-employed, who are not counted in the establishment survey). The household survey is also the source for estimates of the unemployment rate. Individuals are considered unemployed if they were not working but reported being available for work. These two surveys provide alternative perspectives on the total number of workers. The establishment survey measures employment from employers’ point of view, asking businesses how many people they paid for work. The household survey gives workers’ perspective, asking individuals whether they did any paid work. So Where Do Furloughed Federal Workers Fit In? According to BLS, furloughed employees are counted as employed in the establishment survey but are considered unemployed in the household survey. Furloughed employees remain on the federal government’s payroll and, because they will be paid once appropriations are renewed, they continue to accrue wage and salary earnings while on furlough. From the federal government’s perspective, they are employees being paid to do no work, even if that pay is not disbursed on the traditional biweekly schedule. Hence, they are “employed” in the establishment survey. In contrast, federal contractors who are not working and will never be paid for lost work are not counted as employed. In the household survey, furloughed federal employees are considered “unemployed on temporary layoff.” These individuals are not working but are clearly available for work, and so are counted as unemployed. From the worker’s perspective, the only difference between a furlough and conventional unemployment is expectations about the future; in the present, both mean no work and no pay. Consequently, the January jobs report will likely feature an unusual divergence between the change in nonfarm payrolls and the unemployment rate. Which number is right? It depends who you’re talking to.

By Kent Smetters Martin Feldstein was the most influential U.S. economic policy adviser during the past half century. As Chairman of the Council of Economic Advisers under President Reagan, he introduced a new era of supply side economics to a US President, Congress and the nation. When needed, Marty openly disagreed with White House operatives, including issuing public warnings about the negative effects of larger deficits on the economy. The CEA had considerably more independence during those days, prior to President Clinton's introduction of the National Economic Council. Earlier this week, I gave a talk at the Peterson Foundation’s annual Fiscal Summit. I was followed by House Speaker Pelosi whose remarks about her relationship with President Trump naturally made headlines. For policy economists, the real insight, though, went widely unnoticed: Speaker Pelosi acknowledged the value of supply-side economics. While debates about distributional fairness continue, gone are the days where supply-side economics is derided wholesale as merely "trickle down." Supply side economics is now an accepted part of the nation’s policy tool belt; not surprisingly, it sometimes makes more sense to focus on stimulating supply than demand. Marty was the dissertation advisor to scores of students, including me. Incredibly generous with his time, he pushed his students to think hard about the economics intuition of their ideas instead of just hiding behind the math. Yet, despite many of his students taking on successful careers, we always walked in his glow. About 15 years ago, the Mexican government invited Marty and me to give two public lectures to the government, press and various academics. On the plane ride to Mexico City, the overhead television announced Professor Feldstein's imminent arrival. In Mexico City, the speaking auditorium was packed with reporters and television crews from Mexico and various Latin American countries. Marty talked first. When he finished, the press core followed Marty outside of the room, leaving me with a few academics, a couple polite Mexican policymakers, and the echo of my own voice. Several months later, I was on a panel discussion with Marty at a large event in DC. I told the room about the Mexican event and that I had, therefore, arranged with the DC panel organizers (I actually had not) to speak before Marty because I would rather be his warm-up act instead of his closer. I warned the audience that I might end early, and so they risked missing the real event if they left the room and tried to time when Marty would speak. As a political appointee at Treasury during the George W. Bush Administration, the Marty glow was everywhere. Unknown to him (I told him later), virtually every op-ed he wrote was scrutinized in detail within Administration offices and across agencies. I was the second person in Treasury to receive a new type of “phone pager" that could send emails (a monochrome Blackberry), allowing me to engage in these debates with a competitive advantage. Within six months, almost everyone in the Administration had one, and so the debates mostly passed from computer keyboard to fervent clicks on the new phone. Communication technology constantly improves over time, but the teaching of sensible economics stands the course of time, and Marty was a steadfast defender of it to the very end.