Top Income Research Articles

Predicting Poverty

Abstract Poverty prediction models are used to address missing data issues in a variety of contexts such as poverty profiling, targeting with proxy-means tests, cross-survey imputations such as poverty mapping, top and bottom income studies, or vulnerability analyses. Based on the models used by this literature, this paper conducts a study by artificially corrupting data clear of missing incomes with different patterns and shares of missing incomes. It then compares the capacity of classic econometric and machine-learning models to predict poverty under different scenarios with full information on observed and unobserved incomes, and the true counterfactual poverty rate. Random forest provides more consistent and accurate predictions under most but not all scenarios.

Using Survey Data to Estimate Intergenerational Mobility in Income and Education in Portugal

AbstractPrevious studies about intergenerational mobility for the Portuguese economy find that education and income persistence is very high in comparison with other developed economies. We construct relative, absolute, global and local measures of mobility for Portugal, comparing them with existing evidence for this and other countries. These are the intergenerational income elasticity (computed using the two-sample two-stage least squares method), income correlation, rank-rank slope, bottom to top income level probability, the share of individuals earning more than their fathers and also the intergenerational education correlation, the low to high education level probability, and the share of individuals with a higher education level than their fathers. We consider the 1968–1988 cohorts and the 1995 and 2019 waves of the European Community Household Panel and the European Union Statistics on Income and Living Conditions, respectively. Overall, based on the point estimates, women seem to present more mobility in income. Upward income mobility is verified at the bottom while persistence exists at the top. Women present a greater absolute educational mobility. More than 80% of individuals have a higher education than their fathers and full upward education mobility exists for children of low-educated fathers. Mobility in education is higher for the offspring of medium–high-income fathers. Individuals with a high education level, in the medium–high income level or with occupations requiring a higher education level show higher mobility.

Public Healthcare Spending, Business Dynamism, and Wage Inequality: Evidence from Developed and Developing Economies

AbstractThis paper examines the effects of public healthcare spending on wage inequality and social welfare in an economy with rural and urban regions. The objectives are twofold. First, we introduce a simple theoretical framework that captures the link between public healthcare spending and firms' output choice and subsequent wage provisions. We demonstrate that in the short run, a rise in public healthcare services can increase urban output and lower urban unemployment and raise both skilled and unskilled wage rates. In the long run, firm entry produces a business‐dynamism effect that enhances the rate of return on capital, reduces the unskilled wage rate and intensifies wage inequality. Second, we empirically test our theoretical predictions using data from 112 economies in the period 2006–2018. We find evidence supporting our theoretical findings concerning the effects of public healthcare on wage outcomes, especially in the long run. Results are highly consistent across the two groups: advanced economies (mainly OECD/European economies), and emerging and developing economies based on alternative measures for wage inequality, including the GINI coefficient and top income share. The paper provides useful information for the public healthcare policy.

Hospital Transfers for Digit Replants as Cost Saving and Safe While Access to Replantation Procedures Remains Limited.

Finger replantation outcomes are influenced both by injury characteristics and by hospital and patient factors, such as hospital type/location and patient gender or insurance. Finger replantation success rates have been shown to be higher at hospitals with higher volumes of finger replants. This study examines the hospital and patient factors that influence hospital transfer and successful replantation in patients experiencing traumatic finger amputation. A total of 5219 patients were identified in the 2008-2015 National Inpatient Sample (NIS) as having experienced traumatic finger amputation with attempted replantation. Hospital transfer and replant outcomes were compared with variables such as patient demographics and hospital characteristics using χ2 tests, t tests, ANOVA, and logistic regression. Traumatic digit amputation patients were most likely to be transferred to medium or large hospitals in urban areas. Hospital transfer was 1.5 times more likely in White patients than Black or Hispanic patients and 1.6 times more likely in middle income quartile patients than the top income quartile. Postreplant amputation was more likely in patients in the lower three income quartiles, on Medicare, of older age, or with more chronic conditions. Hospital transfer was not associated with changes in the probability of requiring amputation after replantation but was associated with a decreased cost of $5000. Hospital transfers for finger replants are safe with respect to replant failure rates and cost-effective, saving $5000 per procedure. Gaps in equitable access to care remain, warranting further study to improve health equity.

Social disparities in pulmonary embolism and deep vein thrombosis during the coronavirus disease 2019 pandemic from the Nationwide inpatient Sample

ObjectivesStudies have shown that coronavirus disease 2019 (COVID-19) is associated with a hypercoagulable state. Studies have yet to examine the interconnectedness between COVID-19, hypercoagulability, and socioeconomics. The aim of this work was to investigate socioeconomic factors that may be associated with pulmonary embolism (PE), deep vein thrombosis (DVT), and COVID-19 in the United States. MethodsWe performed a 1-year (2020) analysis of the National Inpatient Sample database. We identified all adult patients diagnosed with COVID-19, acute PE, or acute DVT using unweighted samples. We calculated the correlation and odds ratio (OR) between COVID-19 and (1) PE and (2) DVT. We executed a univariate analysis followed by a multivariate analysis to examine the effect of different factors on PE and DVT during the COVID-19 pandemic. ResultsWe identified 322,319 patients with COVID-19; 78,101 and 67,826 patients were identified with PE and DVT, respectively. PE and DVT, as well as inpatient mortality associated with both conditions, are significantly correlated with COVID-19. The OR between COVID-19 and PE was 2.04, while the OR between COVID-19 and DVT was 1.44. Using multivariate analysis, COVID-19 was associated with a higher incidence of PE (coefficient, 2.05) and DVT (coefficient, 1.42). Other factors that were significantly associated (P < .001) with increased incidence of PE and DVT along with their coefficients, respectively, include Black race (95% confidence interval [CI], 1.23-1.14), top quartile income (95% CI, 1.08-1.16), west region (95% CI, 1.10-1.04), urban teaching facilities (95% CI, 1.09-1.63), large bed size hospitals (95% CI, 1.08-1.29), insufficient insurance (95% CI, 1.88-2.19), hypertension (95% CI, 1.24-1.32), and obesity (95% CI, 1.41-1.25). Factors that were significantly associated (P < .001) with decreased incidence of PE and DVT along with their coefficients, respectively, include Asians/Pacific Islanders (95% CI, 0.52-0.53), female sex (95% CI, 0.79-0.74), homelessness (95% CI, 0.62-0.61), and diabetes mellitus (0.77-0.90). ConclusionsIn the Nationwide Inpatient Sample, COVID-19 is correlated positively with venous thromboembolism, including its subtypes, PE and DVT. Using a multivariate analysis, Black race, male sex, top quartile income, west region, urban teaching facilities, large bed size hospitals, and insufficient social insurance were associated significantly with an increased incidence of PE and DVT. Asians/Pacific Islanders, female sex, homelessness, and diabetes mellitus were significantly associated decreased incidence of PE and DVT.

Measuring and taxing top incomes and wealth

Abstract We discuss the measurement of top incomes and wealth in the UK and options for reforming their taxation. First, we highlight the importance of capital gains and migration in understanding long-term trends in top income shares and of survey under-coverage at the top in understanding top wealth shares. We next consider the scope for reforms to the taxation of capital to tackle these inequalities, whilst also improving the efficiency of taxation, emphasizing the roles of capital gains tax, inheritance tax and wealth taxes. Finally, we examine the question of who is taxed, including the tax treatment of highly mobile individuals and of trusts.

Top income inequality and tax policy

Abstract The share of pre-tax income flowing to the top of the UK income distribution increased continually and substantially in the three decades leading up to the financial crisis, but has changed little since 2013. Using microdata sampled from UK tax records, we describe the nature of top incomes in the UK and how they are taxed. We show that wage income is the dominant source of pre-tax income, even for highest-income 0.1% of UK adults. However, ‘active’ business income—derived from self-employment or closely-held incorporated businesses—is considerably more important for the top 1% than for those with lower incomes. High-income wage earners work disproportionately in financial services. The high-income self-employed are predominately working in partnerships in professions such as accountancy and legal services. Overall, UK income taxes are progressive: average tax rates rise with income. Taxes on top incomes have been increased since 2010, with the result that the post-tax share of income flowing to the top has fallen. But average tax rates vary significantly within the top and depend on how income is received. Incomes from business ownership and investment are taxed at lower rates than employment income. We discuss options for reforming the taxation of top incomes.

Discussion of top incomes and tax policy

Abstract This commentary begins with six general observations about top incomes and taxation. Then, it provides some commentary on the article ‘Top income inequality and tax policy’ by Delestre et al. (2024), and concludes with a brief discussion on the revenue potential of top capital gains taxation.

Top incomes and tax policy

Abstract Modern tax systems do not satisfactorily tax the wealthiest individuals. Meanwhile, government revenue needs have increased in the wake of the COVID-19 pandemic. This paper discusses ideas for new revenue sources focused on the top of the distribution: A progressive tax on net wealth with a high exemption threshold, a wealth tax on corporations' stock and a one-off tax on top-end unrealized capital gains. Given the size and concentration of wealth, the revenue potential of these policies is significant.

Top income inequality and tax policy

Top income inequality and tax policy

The welfare state and inequality: were the UK reforms of the 1940s a success?

Abstract This commentary reviews reforms made to the welfare state made in the 1940s by the UK. It also considers education, health and income inequalities in the light of these reforms. The evidence points to more equal access to health care and a reduction in income inequality, especially through a lower share for top incomes. Greater income equality was not, however, mainly a result of redistributive policies. The Beveridge Report was an unsatisfactory blueprint for a social security system, and it did not achieve its objectives of providing adequate benefits while ending means testing. Poverty was reduced but not abolished.

International remittances and labor supply in Nigeria: Do educational attainment and household income matter?

AbstractThe effect of remittances on the labor supply decisions of recipients remains a subject of contention. To contribute to the debate, we investigate how the educational attainment and household income of remittance recipients shape their decisions to work. Using data from the 2018–2019 living standard measurement survey for over 61,000 Nigerians and applying the instrumental variable probit and Tobit techniques, we find that remittances are associated with an occupational switch from agriculture to nonagricultural (paid jobs and nonfarm enterprises) works. Specifically, the results show that remittance recipients are more likely to exit or reduce the hours worked on the farm, regardless of educational attainment and household income status. On the other hand, remittances promote labor supply to paid jobs and nonfarm enterprises, especially among the less educated in Northern Nigeria. In terms of household income, the positive effect of remittances on nonfarm jobs only holds for individuals in the top income quartile, regardless of their region of residence. Our findings are robust to alternative estimation techniques and hold important cues for policymakers.

Tariff wars, unemployment, and top incomes

Motivated by the 2018–19 global tariff war, we develop a multi-country trade model with occupational choice, heterogeneous firms, and unemployment. The model features a complete tariff pass-through and positive optimal tariffs addressing product and labor-market distortions. The quantitative analysis of the model with four countries/regions shows that raising tariffs unilaterally by a country increases welfare but also raises unemployment and top incomes in that country, whereas having the opposite impact on tariff-targeted countries. A global tariff war reduces every country’s welfare, unemployment, and top-income inequality, whereas moving from a worldwide tariff war to free trade raises every country’s welfare, unemployment, and top-income inequality.

Pagetic osteosarcoma: An analysis of the National Cancer Database.

e23503 Background: Pagetic osteosarcoma (Pagetic OS) is a rare cancer typically originating at bones affected by Paget disease. Pagetic OS may present as metachronous multifocal lesions primarily affecting the pelvis, proximal femur, and proximal humerus. There are currently no available sociodemographic descriptions of patients with Pagetic OS. Analyzing sociodemographic factors through the National Cancer Database may unveil important information about its epidemiology and guide future treatment. Methods: A retrospective cohort analysis utilizing the 2004–2020 National Cancer Database (NCDB) included patients with a histologically-confirmed diagnosis of Pagetic osteosarcoma (N = 128) using ICD-O-3 code 9184. Demographic factors (age, sex, race, Hispanic status, educational attainment, insurance status, facility type, distance from facility, and Charles/Deyo score) were analyzed by descriptive statistics. Incidence trends were interpreted in regression analysis. Results: In the NCDB, 128 patients were identified with a confirmed diagnosis of Pagetic osteosarcoma (Pagetic OS) between 2004 and 2020, with a steady incidence of diagnoses per year (R2 = 0.0). The average age of diagnosis was 68.9 years (SD = 13.4, range = 11-90 years). Men were more likely to be diagnosed than women (57% and 43%, respectively). Most patients were non-Hispanic White (71%), insured by Medicare (56%), and lived in a metropolitan area (85%). When compared individually, more patients were in the top median household income quartile (31%). The top two primary sites were the long bones of the lower limb including their associated joints (27%) and the pelvic bones excluding the sacrum, coccyx, and symphysis pubis (27%). Within the cohort, 31 patients received chemotherapy as primary treatment (24%) while 28 patients received primary radiation therapy (22%). Most patients (63%) did not undergo primary site surgery while 39 patients (31%) did receive surgery as primary treatment. The 90-day mortality included 15% of patients, who died 90 or fewer days after surgery was performed. The 2-year survival was 82% while the 5-year survival was 53% and the 10-year survival was significantly lower at approximately 11%. Most individuals (76%) had Charlson-Deyo comorbidity scores of 0. Conclusions: To the best of our knowledge, this NCDB analysis on Pagetic osteosarcoma (Pagetic OS) is the first of its kind on this specific subset of osteosarcoma. Previous research suggests the broader osteosarcoma occurs most frequently in adolescents. This data reveals that Pagetic OS is primarily diagnosed later in life (68.9 years), while still affecting a significantly wide age range (range = 11-90 years). Most patients were non-Hispanic White (71.1%), lived in a metropolitan area (85.2%), and belonged to the top median household income quartile (31.3%). Further research could unveil additional guidance on best practice in diagnosing and treating this cancer.

Dedifferentiated chordoma: Socioeconomic factors and treatment outcomes.

e23507 Background: Dedifferentiated chordoma (DC) is a rare and aggressive chordoma that arises in the spine and skull base characterized by frequent metastasis. Comparatively common in children, these rare tumors arise from the remnants of the fetal notochord. Though surgery and radiation therapy are common treatments, definitive therapeutic guidelines are poorly characterized due to DC’s rarity. Examining the diagnostic trends of DC may yield valuable information about its epidemiology. Thus, the National Cancer Database (NCDB) was analyzed to explore the demographics of patients diagnosed with this rare condition. Methods: The 2004–2020 NCDB recorded 53 patients histologically diagnosed with DC (ICD-O-3 code 9372). Diagnostic trends were analyzed with linear regression and demographic factors such as age, sex, race, and Charles/Deyo score were quantified descriptively. Results: Of the 53 NCDB cases, the most frequently diagnosed group were white (94%), non-Hispanic (85%) women (57%) in the top income quartile (36%) with no comorbidities (87%) living in metropolitan areas with a population of 1 million or more (51%). They often received treatment at academic/research programs (36%) or integrated network cancer programs (21%). In the years 2004-2020, rates of diagnosis remained constant (R2 = 0.0235). Nearly all (98%) of patients were insured, most commonly with Medicare or privately (42% each). The average age at diagnosis was 51 ± 26 years. Over half (55%) of DC cases originated in the pelvic bones, sacrum, or coccyx. Most (74%) patients received surgery, with the most common procedure being wide excision of the primary site (32%), eliminating the tumor (30%). Radiation therapy (55%) was another common treatment. Ninety days post-surgery, 94% of patients were still alive. However, the survivorship (58%) plummets after two years, with an overall survival time of just 5.8 years. Conclusions: This review is the first to explore the demographic and diagnostic factors of DC on the NCDB. Though rare, DC is an aggressive cancer that half of patients succumb to in around 2 years. It commonly originates in the pelvis, sacrum, or coccyx, contrary to prior literature that suggests it affects the skull base or spine. White, non-Hispanic women in the top income bracket were the most frequently diagnosed group. With patient ages ranging from 1 to 86, it affects children and adults alike. Despite this, prior literature mainly explores childhood cases, highlighting a gap in current scientific understanding of DC, a significant barrier to effective treatment. Intriguingly, 51% of DC cases occurred in cities, but 83% of all Americans live in metropolises. This discrepancy suggests that lower-income urban populations may face greater challenges obtaining diagnosis or treatment. Thus, further research into the effects of sociodemographic factors on the prognosis of adult-onset DC is critical to treatment equity for this poorly understood disease.

Analysis of demographic and socioeconomic factors in mixed type rhabdomyosarcoma using NCDB.

e23538 Background: Mixed type rhabdomyosarcoma is a rare soft tissue tumor characterized by the presence of two histologically different subtypes of rhabdomyosarcoma, commonly found in the testis. Despite a majority of the patients undergoing aggressive treatment, the prognosis is poor. Analysis of its diagnostic trends could prove valuable in understanding mixed type rhabdomyosarcoma’s epidemiology. The 2004-2020 National Cancer Database (NCDB) was analyzed to determine the demographic factors of patients with mixed type rhabdomyosarcoma. Methods: A retrospective cohort analysis of the 2004–2020 NCDB, using ICD-O-3 code 8902/3, included patients diagnosed with mixed type rhabdomyosarcoma (N = 94). Descriptive statistics were used to analyze various demographic factors such as sex, age, race, Hispanic status, educational background, insurance status, facility type and location, distance from facility, and Charlson-Deyo score. Incidence trends were examined through regression analysis. Results: The NCDB identified a total of 94 patients with a confirmed diagnosis of mixed type rhabdomyosarcoma from 2004-2020, with a slight decline in incidences diagnosed per year (R² = 0.0903). Females had a slightly higher likelihood of being diagnosed (55%) compared to males (45%), with average age at diagnosis being 40.06 years (SD = 2.8, Range = 1-90). The majority of patients were White (72%) and of non-Hispanic descent (82%). The top primary site was the testis (23%), with endometrium being another major site (15%). Most (83%) patients had a Charles/Deyo comorbidity score of 0. All patients received immunotherapy as primary treatment. Most (46%) patients were privately insured. A majority (26%) of patients received treatment at an academic/research program. A greater number (53%) of patients resided in metropolitan counties with a population larger than 1 million. Most (40%) patients belonged to the highest quartile of income earners of $74,063 or greater. Conclusions: Our research on mixed type rhabdomyosarcoma, the first NCDB analysis of its kind, aims to address a notable knowledge gap on the topic. Similar to information presented in earlier studies, our analysis revealed mixed type rhabdomyosarcoma has a relatively poor prognosis and is mostly observed in White, Non-Hispanic individuals. Analysis on the demographics of this patient population revealed most patients were top quartile income earners, resided in a metropolitan location, and were more likely to be treated at an academic/research program. Further research should aim to better understand the impact of such demographic and socioeconomic features on the diagnosis, treatment plans, and overall mortality rates of patients with mixed type rhabdomyosarcoma.

Demographic and socioeconomic factors of enterochromaffin cell carcinoid.

e16288 Background: Gastric carcinoid tumors, previously known as neuroendocrine neoplasms, are derived from enterochromaffin cells and show an increasing incidence globally. Neuroendocrine tumors often result in hormonal syndromes due to the production of exogenous hormonal peptides by the tumor. Current treatment options include surgical resection and somatostatin pharmacotherapy to blunt tumor growth. A gastric carcinoid malignancy that requires more investigation is enterochromaffin cell carcinoid (ECC). Considering the increasing incidence of gastric carcinoids, a retrospective investigation of demographic data involving enterochromaffin cell carcinoids was performed using the National Cancer Database (NCDB). Methods: The 2004-2019 NCDB was explored to perform a retrospective cohort analysis of patients diagnosed with ECC. 200 patients with a histologically confirmed diagnosis (ICD-O-3 code 8241/3) were identified. Descriptive statistics were used to analyze the following demographic data of patients in this study: race, sex, age, income quartile, facility type, Hispanic status, and Chales/Deyo score. Regression analysis was used to interpret incidence trends. Results: The NCDB identified 200 patients diagnosed with ECC from 2004-2019 with incidence rates varying over this time (R² = 0.0001). The mean age at diagnosis was 55 years (SD = 19.7, range = 14-90). Women (62%) were more likely to be diagnosed than men. The majority of patients were White (86%) and non-Hispanic (80%). Most patients (99.9%) received treatment for the remission of the cancer, predominantly through surgical means (86.5%). More patients were treated in an Comprehensive Community Cancer program (34.5%) than in an academic/research program (26.0%). The majority of individuals lived in metropolitan counties with a population greater than 1 million (54%) than in other metropolitan counties (22%). Most patients were in the fourth quartile of income earners (32.1%). The top primary site was the appendix (40.5%), and most individuals (79.5%) had Charles/Deyo comorbidity scores of 0. Conclusions: To the best of our understanding, this is the first NCDB investigation of Enterochromaffin Cell Carcinoid (ECC) and therefore addresses a meaningful educational disparity on the topic. Most ECC patients tend to be White, non-Hispanic individuals with a primary diagnosis site of the appendix, consistent with previous studies. Upon analyzing the socioeconomic factors of ECC patients, it was found that they often fall into the top income quartile, live in densely populated metropolitan counties, and receive treatment from a comprehensive community cancer program. Due to the significance of these demographic and socioeconomic factors, more research is needed to understand their impact on the diagnosis and treatment of ECC.

The demographics of Sertoli cell carcinoma: A National Cancer Database analysis.

e17012 Background: Sertoli cell tumors (SCTs) are an extremely rare and lethal diagnosis responsible for roughly 1% of all testicular and ovarian carcinomas. No uniform treatment regimen with sufficient results has been established. Current combination of retroperitoneal lymph node dissection, chemotherapy, and radiation therapy provides the best known outcome. Analyzing trends in diagnosis may provide valuable information about its epidemiology. The National Cancer Database (NCDB) was analyzed to establish the demographic factors in diagnosed patients. Methods: A retrospective cohort study from 2004 to 2020 was conducted using the NCDB. 159 patients were found with a histologically confirmed diagnosis of SCTs (ICD-8640-3). Race, gender, Hispanic origin, type of treatment facility, insurance coverage, and distance from patient residence to treatment facility were analyzed via descriptive statistics, and trends in incidence were interpreted using regression analysis. Results: The number of patients diagnosed with SCTs has remained relatively consistent since 2004 (R^2 = 0.07), with approximately 9 diagnoses per year. Of the 159 patients found, the average age of diagnosis was 44.6 years (SD = 19.1, range = 2 – 89 years). A majority of patients were male (67%), while the remaining third were female (33%). Most of the patients were White (73%). Patient Hispanic status could not be reliably determined due to limited data. More patients were in the top income quartile (31%) than the second (24%), third (18%), or fourth (14%) quartiles, with 13% having unknown income status. Most patients were privately insured (59%), followed by coverage under Medicare/Medicaid (28%). More patients lived in metropolitan counties with a population of 250,000 or more (76%) compared to less-populated urban or rural regions (24%). The majority received treatment at comprehensive community cancer programs or academic/research programs (43%). Patients traveled an average of 30.2 miles (SD = 64.0, range = 0.5 – 474.9) to reach the treatment facility. Conclusions: SCTs remains a rare diagnosis. This study pursues a significant gap in existing knowledge on patient characteristics that may present with SCTs. Most patients were White, privately insured, in the top income quartile, and lived in metropolitan areas. This novel study may contribute data towards understanding the relationship between patient socioeconomic status and access to diagnosis. Further research is required to uncover the role of socioeconomic status in SCTs.

Forecasting the UK top 1% income share in a shifting world

AbstractUK top income shares have varied hugely over the past two centuries, ranging from more than 30% to less than 7% of pre‐tax national income allocated to the top 1 percentile. We build a congruent dynamic linear regression model of the top 1% income share allowing for economic, political and social factors. Saturation estimation is used to model outliers and trend breaks, proxying underlying structural changes driving income inequality in the UK. We use the model to forecast the top 1% income share over the last 15 years, and compare to a range of forecast devices. Despite a well‐specified constant parameter model conditioning on significant explanatory variables, the best performing forecasts are obtained from a random walk and a smoothed random walk. These results are explained by the presence of shifts in the income share over the forecast period, resulting in forecasts from equilibrium correction models converging to the wrong equilibrium. Our best prediction for 2026 based on the most recent data from 2021 (a 5‐year ahead projection) is that the pre‐tax top 1% income share will remain at the most recent realized value of 12.7%, but there is a large degree of uncertainty, with a 95% confidence band ranging from 10% to 15.7%.

Political Language in Economics

Abstract Does academic writing in economics reflect the political orientation of economists? We use machine learning to measure partisanship in academic economics articles. We predict the observed political behaviour of a subset of economists using phrases from their academic articles, show good out-of-sample predictive accuracy and then predict partisanship for all economists. We then use these predictions to examine patterns of political language in economics. We estimate journal-specific effects on predicted ideology, controlling for author and year fixed effects, that accord with existing survey-based measures. We show considerable sorting of economists into fields of research by predicted partisanship. We also show that partisanship is detectable even within fields, even across those estimating the same theoretical parameter. Using policy-relevant parameters collected from previous meta-analyses, we then show that imputed partisanship is correlated with estimated parameters, such that the implied policy prescription is consistent with partisan leaning. For example, we find that going from the most left-wing authored estimate of the taxable top income elasticity to the most right-wing authored estimate decreases the optimal tax rate from 77% to 60%.