ALASKA STATE LEGISLATURE  HOUSE SPECIAL COMMITTEE ON WAYS AND MEANS  April 26, 2023 6:03 p.m. MEMBERS PRESENT Representative Ben Carpenter, Chair Representative Jamie Allard Representative Tom McKay Representative Kevin McCabe Representative Cathy Tilton Representative Andrew Gray Representative Cliff Groh MEMBERS ABSENT  All members present COMMITTEE CALENDAR  PRESENTATION(S): ECONOMIC IMPACT OF REVENUE MEASURES - HEARD PREVIOUS COMMITTEE ACTION  No previous action to record WITNESS REGISTER ALEXEI PAINTER, Director Legislative Finance Division Legislative Affairs Agency Juneau, Alaska POSITION STATEMENT: Co-presented the PowerPoint, titled "Economic Impact of Revenue Measures." CONOR BELL, Fiscal Analyst Legislative Finance Division Legislative Affairs Agency Juneau, Alaska POSITION STATEMENT: Co-presented the PowerPoint, titled "Economic Impact of Revenue Measures." ACTION NARRATIVE 6:03:31 PM CHAIR BEN CARPENTER called the House Special Committee on Ways and Means meeting to order at 6:03 p.m. Representatives Allard, McKay, McCabe, Tilton, Gray, Groh, and Carpenter were present at the call to order. ^Presentation(s): Economic impact of Revenue Measures Presentation(s): Economic impact of Revenue Measures    6:04:14 PM CHAIR CARPENTER announced that the only order of business would be the presentation on the economic impacts of revenue measures. CHAIR CARPENTER informed members that the presentation provides follow-up responses to questions posed in a previous meeting. 6:04:58 PM ALEXEI PAINTER, Director, Legislative Finance Division, Legislative Affairs Agency, co-presented the PowerPoint, titled "Economic Impact of Revenue Measures" [hard copy included in the committee packet]. He began on slide 2, listing the presentation's topics, as follows: the comparison of sales and income tax revenue projections; distributional analysis concerning the impact of broad-based taxes and how the PFD changes by income; and the regional analysis and its limitations. He stated that the Legislative Finance Division is not advocating for any fiscal plan or measure. 6:06:02 PM CONOR BELL, Fiscal Analyst, Legislative Finance Division, Legislative Affairs Agency, co-presented the PowerPoint, titled "Economic Impact of Revenue Measures." He read from slide 3, which read as follows [original punctuation provided]: Department of Revenue Chainbridge, which has micro-data for other states, developed estimates for four sales tax proposals based on 2019-2020 data • DOR adjusted for nationwide retail sales growth • Overall results are similar to past DOR models when adjusted for nationwide retail sales growth since previous modeling. Chainbridge data allows for more detailed modeling of exemptions • Assumes revenue growth of 3% per year (2.5% inflation plus 0.5% population growth) • Buckeye Institute Adjusts DOR 2016 HB 5004 fiscal note for impacts on consumer and industry behavior (18% revenue reduction) • Tax Foundation Uses BEA Personal Consumption Expenditures data to estimate tax base and value of exemptions. Assumes 85% collections rate. 6:09:21 PM MR. BELL, in response to a question from Chair Carpenter, stated that the most recent Department of Labor population projections relate that the trend of population decline is reversing. 6:10:20 PM MR. BELL returned to the presentation on slide 4. He read from the the slide, which read as follows [original punctuation provided]: • Department of Revenue Uses 2020 IRS Statistics of Income data with 5% nonresident adjustment • Scaled up ~20% to account for wage growth since 2020 • Assumes revenue growth of 3% per year (2.5% inflation plus 0.5% population growth) • Institute on Taxation and Economic Policy Methodology similar to DOR, but uses 2018 IRS Statistics of Income data • Buckeye Institute Internal static model with similar results to 2018 DOR estimate. Adjusts for impacts on consumer and industry behavior. 6:12:15 PM MR. BELL moved to and read slide 5, which read as follows [original punctuation provided]: • Dynamic modeling estimates how a policy change impacts individual and business behavior • Buckeye is the only publicly-available source with Alaska-specific dynamic modeling on revenue options • Buckeye's dynamic modeling estimates sales tax revenues decline by 18% when accounting for behavior changes • DOR's tax projections have always been static models due to the difficulty of objectively predicting taxpayer behavior. The degree to which company behavior is influenced by tax changes has been central to past oil tax debates in the legislature, for example. 6:13:36 PM MR. BELL, in response to a question from Chair Carpenter, stated that modeling accounts for both taxes. He explained that, through applying dynamic modeling, the flat income tax assumption is lowered by 15 percent. He said a progressive income tax with a top marginal rate of 7 percent shows a 43 percent decline in revenues collected. 6:14:22 PM MR. BELL, in response to a question from Representative McCabe, explained that an income tax would cause a 43 percent decrease in revenue, based on the estimates by Buckeye Partners. He said part of this is from high-income earners having more incentive to decrease their tax liability. REPRESENTATIVE MCCABE offered the understanding that Mr. Bell's comments mean the high-income earners would have a greater ability to move out of state; therefore, they would not pay the tax. MR. BELL responded that Buckeye Partners' analysis suggests there would be a greater change in behavior among the high- income earners. 6:16:03 PM REPRESENTATIVE GROH commented that 40 states have income taxes, and Alaska is the only state which does not have a broad-based tax; more than two-thirds of states have both a sales tax and income tax. He asked Mr. Bell if the Buckeye Partners' analysis seemed right to him. CHAIR CARPENTER added to the question, inquiring whether there is data which supports or refutes the analysis. 6:17:04 PM MR. PAINTER answered that the 15 percent to 18 percent range is more standard than 43 percent; however, he expressed doubt whether there is an objective data review of what would be a generally accepted number. He expressed the opinion that 43 percent is unusually high. CHAIR CARPENTER concurred. He said there are several states which have high income taxes and have experienced years of outward migration. 6:18:20 PM REPRESENTATIVE GRAY expressed surprise that any state which instituted a progressive income tax would have 43 percent of the higher earners move out of state. 6:18:54 PM REPRESENTATIVE MCCABE expressed the opinion that the conversation is bigger than just 43 percent of people leaving the state. He suggested that people find ways to shelter money and remove it from the revenue stream. 6:20:15 PM CHAIR CARPENTER suggested that the committee acknowledge there would be a behavior change which may not be positive for state revenue. 6:20:52 PM MR. BELL returned to the PowerPoint Presentation to read slide 6, which read as follows [original punctuation provided]: Impacts to hypothetical families of four with annual income of $25,000, $50,000, $100,000 or $200,000 Compares three policy changes, each with roughly $640 million revenue impact: 4% broad-based sales tax exempting groceries 1.9% income tax on Adjusted Gross Income (AGI) with no deductions $1,000/person PFD reduction Estimates use various simplifying assumptions and are for illustrative purposes only MR. BELL explained that the estimates in the chart on slide 6 are rough approximations. He noted that the 4 percent sales tax would have a higher impact on lower income households because these households tend to spend a greater portion of income each month. He continued that with an income tax, the impact would be the same for families earning $25,000, $50,000, and $100,000 annually, while families earning $200,000 annually would pay a slight bit lower rate because of the assumption that these families would itemize the tax against their federal tax liability. He said the $1,000 permanent fund dividend (PFD) reduction has a higher impact on lower income families, assuming that the PFDs are subjected to federal income tax at the families' marginal rate. 6:23:22 PM CHAIR CARPENTER asked about the reduction on the chart as it relates to the PFD. 6:23:59 PM MR. PAINTER explained that about eight years ago there was not much data available, so the Legislative Finance Division investigated how a distributional analysis could be done in a way which is easily understandable. He said an analysis was published using hypothetical families of various sizes. 6:25:13 PM MR. BELL returned to the presentation to read slide 7, which reads as follows [original punctuation provided]: Policy design can significantly impact the distributional impact of a broad-based tax For example, adding a deduction to an income tax or a progressive rate structure would change the distributional impact of the tax • Sales tax exemptions can also impact the distributional impact because of the varying mix of goods and services purchased at different income levels. 6:26:08 PM MR. BELL moved to and read slide 8, which read as follows [original punctuation provided]: • A 2016 study by ISER economists analyzed the distributional impacts of revenue changes, including indirect impacts • The chart below shows the impact on disposable income per $100 million raised, not the total impact of the policy changes listed • Note that the study predates major federal income tax changes from the 2017 Tax Cuts and Jobs Act MR. BELL said the table from the 2016 Institute of Social and Economic Research (ISER) University of Alaska Anchorage study is attached. He explained that the chart shows the impact on disposable income per $100 million in revenue raised. CHAIR CARPENTER pointed out that, from the 50th to 90th percentile, the different options are equally distributed. He said the committee is looking at multiple options, and the chart shows an either/or option between different tax proposals. He asked what the distribution analysis would be if the PFD cut was not applied, and a sales tax was implemented instead. He asked whether a large negative distributional analysis at the lower levels with the PFD exists, and if it is no longer there, should it be added into the increase on one of the other tax options. 6:29:10 PM MR. PAINTER responded that Chair Carpenter's comments highlight the limitations of the analysis, in that any option to close the deficit is going to have an impact on the economy, as well as to private incomes. He said that spending $1 billion from the constitutional budget reserve (CBR) is not sustainable in the long term but would also not have an impact on the economy, whereas taking $1 billion from the economy would have an impact. He said the challenge of doing distributional analysis using the past is determining the comparison. 6:31:02 PM REPRESENTATIVE MCCABE asked whether there would be an impact since the PFD has a multiplier effect of private money on the economy. MR. PAINTER answered yes and said the 2016 ISER study examined the short-term economic impacts of revenue and budget options, attempting to get an approximation of the different options to close the deficit. He said that all the options on the chart affect the economy in some way, and the questions become: what is the multiplier on the economy and what is the distribution analysis? 6:32:41 PM REPRESENTATIVE GROH asked if there are other differences between the ISER mythology and the division's mythology, as seen on slide 6 and slide 8. MR. BELL responded that the ISER study is more thorough. 6:33:40 PM MR. PAINTER returned to the presentation, and paraphrased slide 9, which read as follows [original punctuation provided]: Any fiscal policy change will be felt differently across Alaska • Due to limited data, it is difficult to estimate regional differences An income tax and PFD are relatively easy to compare across geographic areas because of the existence of detailed income data Sales taxes are difficult to compare across geographic regions because of differences in consumption patterns • The next several slides show different ways of illustrating regional differences, though all have limitations • Alaska Economic Trends' annual Cost of Living issue provides more detail on comparing costs across communities 6:35:55 PM MR. PAINTER moved to slide 10 to show the per-capita personal income by region. He said the Bureau of Economic Analysis has data from 2021 which shows the average amount of income for every person in any given area. He advised that the data shown on the slide are estimates at an individual level, not at the household level, and 2021 was unusual because of the government aid during the pandemic. He said the analysis does not account for differences in cost of living, distribution of income, or the significance of noneconomic activities. He said that analyses cannot ascertain how a two percent sales tax would impact residents differently because of differences in lifestyles and economic activity; therefore, it is difficult to understand the true impact on the economy and individuals. He said that the division could apply an income or PFD reduction and see a percentage which would apply to the median income in various regions. 6:37:28 PM REPRESENTATIVE MCKAY brought attention to the 2021 per-capita personal income figure for Bristol Bay, which is listed as $155,155. MR. PAINTER answered that this is because of the [commercial] fishing activity in Bristol Bay. 6:37:52 PM MR. PAINTER moved to slide 11 and explained that the U.S. Department of Defense (DoD) tracks the cost of living in multiple Alaska communities in order to account for the cost-of- living differences for service members stationed outside the continental U.S. He said the index excludes housing and may not reflect consumption habits of residents. He explained that the average index in the U.S. is 100, while the highest in Alaska is in Bethel and Cordova at 146. He pointed out that this is 46 percent higher than the average U.S. city. He stated that the lowest index in Alaska is Anchorage and Wasilla at 128. 6:40:28 PM CHAIR CARPENTER pointed out that there are portions of the state with a higher cost of living, like communities that are on the road network, while those not on the network have a lower cost of living. MR. PAINTER explained that DoD is also looking at the amount of on-base expenditures and off-base expenditures. He said that a community like Kodiak, which has a large coastguard base, would presumably have more on-base shopping available at a reduced cost. 6:41:36 PM REPRESENTATIVE GRAY shared that the commissary at the Joint Base Elmendorf Richardson (JBER) is not much cheaper than most grocery stores in Anchorage. He asked about income data in Bethel, as Bristol Bay is at $155,155, but Bethel is the lowest in per-capita income. He suggested that the data does not include the villages off the road system, as there are typically higher prices in such communities. MR. PAINTER said there is no data on this, and, while there is significant information about the hub communities, it is not representative of the much smaller communities. He reiterated that Bristol Bay has [commercial] fishing activity; therefore, income inequality may be affecting the data. He pointed out that Bethel may have a difference in consumption patterns since more of the economy is dependent on subsistence. 6:44:23 PM REPRESENTATIVE GROH emphasized that communities shown in the cost-of-living chart are much bigger than the villages not shown, and remote places would have a higher cost of living. MR. PAINTER responded that smaller communities are going to have higher prices and would be impacted. He said that the remote communities do not have a military presence, so the data is not available. REPRESENTATIVE GROH commented that the income and sales tax estimations are not representative of the proposals in front of the committee. MR. PAINTER explained that if the division were to estimate the impact of the bills before the committee, there would be data limitations as well. He explained that there is no distributional analysis on sales tax in Alaska. He said that data from Minnesota was used instead as it has sophisticated data and analysis. 6:46:12 PM REPRESENTATIVE ALLARD expressed the understanding that some subsistence would apply. She questioned whether other subsidies would be accounted for. MR. PAINTER expressed the understanding that the data on income includes all cash income, but the value of noneconomic sources, like subsistence, would not be accounted for. REPRESENTATIVE ALLARD asked about public assistance. 6:47:12 PM MR. BELL said that government transfer payments are included. 6:47:28 PM REPRESENTATIVE TILTON asked if income includes investment and corporate dividends. MR. BELL answered yes. 6:47:56 PM MR. PAINTER returned to the presentation on slide 12 and showed a graph of fuel costs by area. He said the data is sourced from the Alaska Department of Commerce, Community, and Economic Development's report from July 2022, titled Current Community Conditions: Fuel Prices Across Alaska. He noted that the North Slope Borough subsidizes heating fuel, setting the current price at $1.50 a gallon; however, gasoline is more expensive because it is not subsidized, and the current price is at $7.20 a gallon. He expressed uncertainty as to why heating fuel is cheap in King Cove. Looking at urban areas, he said, fuel is typically priced at the $3 a gallon to $4 a gallon range, with prices in rural areas being about double this. 6:49:21 PM MR. PAINTER moved to slide 13 which showed a graphic depicting wild food harvests by area in 2017. He commented that it is difficult to quantify the impact on people because of the differences in noncash sources available. He explained that the Anchorage harvest is about 50 pounds per person, whereas in arctic Alaska it is over 400 pounds per person. He said it is difficult to examine prices alone and turn this into an idea of how a sales tax could potentially impact different communities because of the vast differences in the economies. 6:50:44 PM MR. PAINTER moved to slide 14, which read as follows [original punctuation provided]: 'Tax base' is the estimate of total economic activity subject to a given tax • Residents in rural areas will often purchase goods in hub communities, making geographic comparisons difficult • Visitors and seasonal workers also can impact the amounts MR. PAINTER said that residents in rural areas will often buy goods in hub communities, further making geographic comparisons difficult. He talked about the dynamics of hub communities and sales tax. He said a person who lives in Haines may do shopping in Juneau, or on the internet. He stressed the difficulty when considering different types of consumption and the impact of a sales tax. He said that another factor would be visitors and seasonal workers. 6:54:51 PM REPRESENTATIVE GRAY asked about Hydaburg and why the tax base is lower per capita. MR. PAINTER offered his understanding that Hydaburg is not on the Prince of Wales road system and individuals opt to travel to Ketchikan; however, Thorne Bay is on Prince of Wales and is an example of a mini hub which bounces off other hubs, such as Ketchikan and Juneau. He added that Hydaburg may just have one store. 6:56:25 PM REPRESENTATIVE MCCABE noted that there are many communities with no stores, so people do air taxi runs. He said that typically a community will buy supplies a month at a time. REPRESENTATIVE GRAY pointed out that the costs are fifty times more in Unalaska. He noted that Unalaska has seasonal workers. 6:57:56 PM REPRESENTATIVE GROH asked the presenters to address personal income in different areas and to point out the average versus the median. MR. PAINTER stated that per capita would be an average, meaning the entire income of an area is divided by the number of earning individuals, whereas a median income would be a middle fiftieth percentile of earning individuals. He said that some communities, like Bristol Bay, may have [commercial] fishing vessels and more per-capita income, dragging the number up, where comparatively the median earner may be earning less than $155,155 a year. REPRESENTATIVE GROH asked about the uncertainty of the data presented and how this relates to the uncertainty concerning the difference between an income tax versus a sales tax. MR. PAINTER answered that in distributional analysis using income tax there is a small degree of uncertainty because the Internal Revenue Service publishes data by income level, and this makes it easy to determine the amount collected and how Alaskans fall into different income brackets; however, he said that a sales tax is harder to ascertain since there is no quality statewide data available. 7:02:42 PM CHAIR CARPENTER thanked the presenters. 7:03:03 PM ADJOURNMENT  There being no further business before the committee, the House Special Committee on Ways and Means meeting was adjourned at 7:03 p.m.