The Unequal Gains From Product Innovations: Evidence From .
The Unequal Gains from Product Innovations:Evidence from the US Retail SectorXavier Jaravel, London School of EconomicsSeptember 21, 2017
IntroductionWho benefits from innovation?IIncome channel: extensive literature on skill-biased technical change[Acemoglu 1998, Goldin and Katz 1998, Autor, Levy and Murnane 2003]IExpenditure channel: new products can affect purchasing-power acrossincome groups directly (by targeting specific groups) and indirectly(through competition with existing products)This paper investigates the impact of product innovations oninequality through the expenditure channelIITheory:Shifts in income distribution Z Increased demand for premium productsZ Shift in direction of product innovationsZ Increase in purchasing-power inequalitySeveral empirical tests support this theory, primarily using scanner datain US retail sectorThis has implications for inflation inequality and the price indexationof certain government programs1
IntroductionWho benefits from innovation?IIncome channel: extensive literature on skill-biased technical change[Acemoglu 1998, Goldin and Katz 1998, Autor, Levy and Murnane 2003]IExpenditure channel: new products can affect purchasing-power acrossincome groups directly (by targeting specific groups) and indirectly(through competition with existing products)This paper investigates the impact of product innovations oninequality through the expenditure channelIITheory:Shifts in income distribution Z Increased demand for premium productsZ Shift in direction of product innovationsZ Increase in purchasing-power inequalitySeveral empirical tests support this theory, primarily using scanner datain US retail sectorThis has implications for inflation inequality and the price indexationof certain government programs1
IntroductionWho benefits from innovation?IIncome channel: extensive literature on skill-biased technical change[Acemoglu 1998, Goldin and Katz 1998, Autor, Levy and Murnane 2003]IExpenditure channel: new products can affect purchasing-power acrossincome groups directly (by targeting specific groups) and indirectly(through competition with existing products)This paper investigates the impact of product innovations oninequality through the expenditure channelIITheory:Shifts in income distribution Z Increased demand for premium productsZ Shift in direction of product innovationsZ Increase in purchasing-power inequalitySeveral empirical tests support this theory, primarily using scanner datain US retail sectorThis has implications for inflation inequality and the price indexationof certain government programs1
Motivating Example: Cost of Detergent (per 100 Loads)BEFOREMoreAFTER 10HIGHERPRICEAll LiquidUPC 9 53228 02121 9 5LOWERPRICEAll PowderUPC 7 74205 55160 42
Motivating Example: Cost of Detergent (per 100 Loads)BEFORE 10MoreAFTER 21 8AFTERHIGHERPRICEAll LiquidUPC 9 53228 02121 9 5Tide PodsAll LiquidUPC 8 2218 00201 5UPC 9 53228 02121 9 5LOWERPRICEAll PowderUPC 7 74205 55160 4All PowderUPC 7 74205 55160 42
Motivating Example: Cost of Detergent (per 100 Loads)MoreAFTERBEFORE 10 21 8AFTERHIGHERPRICEAll LiquidUPC 9 53228 02121 9 5Tide PodsAll LiquidUPC 8 2218 00201 5UPC 9 53228 02121 91 5LOWERPRICEAll PowderUPC 7 74205 55160 4All PowderUPC 7 74205 55160 41 Increased demand2
Motivating Example: Cost of Detergent (per 100 Loads)MoreAFTERBEFORE 10 21 82 AFTERHIGHERPRICEAll LiquidUPC 9 53228 02121 9Tide PodsAll LiquidUPC 8 2218 00201 5UPC 9 53228 02121 91 5 5LOWERPRICEAll PowderUPC 7 74205 55160 41 Increased demandAll PowderUPC 7 74205 55160 42 Increased entry2
Motivating Example: Cost of Detergent (per 100 Loads)MoreAFTERBEFORE 10 213 82 AFTERHIGHERPRICEAll LiquidUPC 9 53228 02121 9Tide PodsAll LiquidUPC 8 2218 00201 5UPC 9 53228 02121 91 5 5LOWERPRICEAll PowderUPC 7 74205 55160 41 Increased demand2 Increased entryAll PowderUPC 7 74205 55160 43 Increased price competition2
Main FindingsIn retail sector (2004-2015), higher-income householdsexperienced a faster increase in product variety and lowerinflation on continued productsIAnnual inflation was 65 basis points lower for households earning above 100k vs. below 30kThis was largely due to the supply response to changes indemand induced by shifts in the income distributionIResearch design in two steps:FFIdentify effect of demand on supply using changes in age and incomedistributions over time as demand shiftersApply point estimates to changes in demand induced by shifts in USincome distributionIAccounts for over 80% of inflation differenceISimple model rationalizes evidence (endogenous entry and markups)3
Related LiteraturesLiterature on innovation and inequalityIFactor-augmenting technical change: Goldin and Katz (1998), Acemoglu(1998, 2002, 2007), Krusell, Ohanian, Rios-Rull and Violante (2000), Greenwoodand Yorukoglu (1997), Galor and Moav (2000), Garicano and Rossi (2004)ISector-augmenting technical change: Acemoglu and Linn (2004), Acemoglu,Aghion, Bursztyn and Hemous (2012), Boppart and Weiss (2013) and Comin,Lashkari and Mestieri (2016)IIProduct cycle: Schumpeter (1942), Vernon (1966), and Matsuyama (2002)Contribution: show theoretically and empirically the implications ofendogenous innovations across product space for inequalityLiterature on inflation inequalityIExtensive literature investigating inflation experiences of differenthousehold groups: Amble and Stewart (1994), Garner, Johnson and Kokoski(1996) and Hobjin and Lagakos (2003), Murphy and Garvey (2004), Chiru (2005),McGranahan and Paulson (2005)IRecent work measuring inflation inequality using scanner data: Brodaand Romalis (2009), Argente and Lee (2016), Kaplan and Schulhofer-Wohl (2016)IContribution: show long-term trend of inflation inequality in scannerdata (not business-cycle phenomenon) and importance of aggregation bias 4
Related LiteraturesLiterature on innovation and inequalityIFactor-augmenting technical change: Goldin and Katz (1998), Acemoglu(1998, 2002, 2007), Krusell, Ohanian, Rios-Rull and Violante (2000), Greenwoodand Yorukoglu (1997), Galor and Moav (2000), Garicano and Rossi (2004)ISector-augmenting technical change: Acemoglu and Linn (2004), Acemoglu,Aghion, Bursztyn and Hemous (2012), Boppart and Weiss (2013) and Comin,Lashkari and Mestieri (2016)IIProduct cycle: Schumpeter (1942), Vernon (1966), and Matsuyama (2002)Contribution: show theoretically and empirically the implications ofendogenous innovations across product space for inequalityLiterature on inflation inequalityIExtensive literature investigating inflation experiences of differenthousehold groups: Amble and Stewart (1994), Garner, Johnson and Kokoski(1996) and Hobjin and Lagakos (2003), Murphy and Garvey (2004), Chiru (2005),McGranahan and Paulson (2005)IRecent work measuring inflation inequality using scanner data: Brodaand Romalis (2009), Argente and Lee (2016), Kaplan and Schulhofer-Wohl (2016)IContribution: show long-term trend of inflation inequality in scannerdata (not business-cycle phenomenon) and importance of aggregation bias 4
SummaryIn retail, inflation was much lower forhigher-income households.because supply responds to changes indemand.induced by shifts in the income distribution.5
SummaryIn retail, inflation was much lower forhigher-income households.because supply responds to changes indemand.induced by shifts in the income distribution.5
SummaryIn retail, inflation was much lower forhigher-income households.because supply responds to changes indemand.induced by shifts in the income distribution.5
Roadmap1Data2Inflation across Income Groups3The Response of Supply to Market Size Effects6
Roadmap1Data2Inflation across Income Groups3The Response of Supply to Market Size Effects6
Scanner DataNielsen Homescan Consumer Panel[Aguiar & Hurst 2007, Einav, Leibtag &Nevo 2008, Broda & Romalis 2009, Broda & Weinstein 2010, Stroebel & Vavra 2014 ]IHouseholds scan prices and quantities for products with barcodes soldin US from 2004 to 2013 (e.g in department/grocery/drug/convenience stores)IHousehold characteristics: income, age, education, occupation,MSA, composition, .IRepresentative of 40% of household expenditures on goods,15% of total household expenditures More on Consumptions Baskets7
ANDISEBEAUTYAND HEALTHALCOHOLFOOD8
RCHANDISEBEAUTYAND HEALTHALCOHOLFOOD8
Roadmap1Data2Inflation across Income Groups3The Response of Supply to Market Size Effects9
Roadmap1Data2Measuring Inflation across Income Groups31Price changes for continued products (90% of spending)2Valuing new and exiting products3Aggregation bias4Evidence outside retailThe Response of Supply to Market Size Effects10
Price Changes for Continued ProductsDifferent price indices put different weights on the product-level pricechanges (substitution):nLaspeyres Index : P L pt puu0 su0u 1CES Exact Price Index : PCES Πcputpu0 wutwith put price, sut spending share and wut Sato-Vartia (1976) weights.Compute separate price indices across income groupsIIn baseline result: three income groups, price index is nested CES,and product u is a UPC10
Nested CES Inflation Rate, 2004-2015(Annualized, %)Price Changes for Continued Products21.81.61.41.2Below 30,000With More Income Groups 30,000 to 100,000Household IncomeAcross DepartmentsAcross YearsBy Age-Income GroupsAbove 100,000Additional Checks11
ce Between Annualized Average InflationRates (pp) for Households with Income Below 30k and Housholds with Income Above 100kNo Differential Substitution Effects.85.8.75.7.65.55.6.45.5.412
Roadmap1Data2Inflation across Income Groups31Price changes for continued products2Valuing new and exiting products3Aggregation bias4Evidence outside retailThe Response of Supply to Market Size Effects13
Specifications0.65Nested CES, Elasticity 2.09from Handbury (2013)0.65Nested CES, Elasticity 4from Dube et al (2005)0.65Nested CES, Elasticity 7from Montgomery and Rossi (1999)0.65Nested CES, Elasticity 11.5from Broda and Weinstein (2010)0.65Nested CES with Hausman (2003)Approximation for New Products,Estimated Elasticities (Median of 6.5)0.65Nested CES-Translog, EstimatedSemi-Elasticities (Median of 1.06)0.62SpecificationNested CES, Estimated Elasticities(Median of 6.5)0.781.150.840.740.700.680.640.51Inflation Difference between High- and Low-Income Households,2004-2013 (Annualized, pp)Continued ProductsDetailed Estimation ResultsProduct Variety13
Roadmap1Data2Measuring Inflation across Income Groups31Price changes on continued products2Valuing new and exiting products3Aggregation bias4Evidence outside retailThe Response of Supply to Market Size Effects14
Decomposition of Inflation DifferenceClassifying products into categories indexed by C , the inflationdifference between high- and low-income can be decomposed as:[Diewert 1975]!πH πL (sCH sCL )πCC {zBetween sC (πCH πCL )C} {zWithin}with sCi share of spending of income group i on C ,πCi the inflation experienced by income group i on C ,π C average inflation rate in C ,sC average spending share in C .Conduct decomposition for various levels of aggregation, using thenested CES price index for continued products14
Aggregation Bias“Between” decomposition:Aggregation Level(Broad to Narrow)DepartmentShare of InflationDifference Explained (%)8.6(e.g. fresh produce vs. health and beauty care)Product Group21.4(e.g. deodorant vs. hair care)Product Module42.8(e.g. men’s vs. women’s hair coloring)More Decomposition ResultsThis explains why old literature has found much smaller inflationinequality [Hobijn & Lagakos 2003, McGranahan & Paulson 2005, Chiru 2005]IContrast with recent literature on inflation using scanner data:Argente and Lee (2016), Kaplan and Schulhofer-Wohl (2016)15
Aggregation Bias“Between” decomposition:Aggregation Level(Broad to Narrow)DepartmentShare of InflationDifference Explained (%)8.6(e.g. fresh produce vs. health and beauty care)Product Group21.4(e.g. deodorant vs. hair care)Product Module42.8(e.g. men’s vs. women’s hair coloring)More Decomposition ResultsThis explains why old literature has found much smaller inflationinequality [Hobijn & Lagakos 2003, McGranahan & Paulson 2005, Chiru 2005]IContrast with recent literature on inflation using scanner data:Argente and Lee (2016), Kaplan and Schulhofer-Wohl (2016)15
Roadmap1Data2Measuring Inflation across Income Groups31Price changes on continued products2Valuing new and exiting products3Aggregation bias4Evidence outside retailThe Response of Supply to Market Size Effects16
Evidence Outside RetailUse CPI and CEX data to assess patterns outside retail:[McGranahan &Paulson 2005]IPrice series on 48 expenditure categories going back to 1953,covering full consumption basketIUsing expenditure shares fixed at 1980-1985 levels, compute inflationfor baskets of households in top vs. bottom income quintilesISubject to aggregation bias, but still useful16
Long-Term Inflation 019751970196519619501195Relative Price Index for Baskets ofBottom versus Top Income Quintiles1.15YearRelative price index is normalized to 1 in 1953.Laspeyres inflation rates are computed using 1980-1985 expenditure shares.RobustnessTFP AnalysisPatent Analysis17
Implications for InequalityOver 2004-2015, nominal increase in food stamp benefits should havebeen 31.4% (instead of 23.2%) to preserve purchasing powerFrom CEX, spending shares in (Nielsen) retail for top and bottomincome quintiles are:α Q1 18%α Q5 12%Under Cobb-Douglas upper nest, change in purchasing-powerinequality per year over 2004-2015 given by: log(Y Q1 ) log(Y Q5 ) α Q1 log(PQ1 ) α Q5 log(PQ5 ) {z} {z}Income: 0.93 ppRetail Inflation: 0.22 pp eQ1 ) (1 α Q5 ) log(PeQ5 ) (1 α Q1 ) log(P {z}Inflation Outside Retail 018
Roadmap1Data2Inflation across Income Groups3The Response of Supply to Market Size Effects19
Descriptive EvidenceProduct modules that grow faster characterized by:IFaster increase in product varietyIIncreasing competition between manufacturersILower inflation on continued productsIMore spending from high-income householdsGraphGraphGraphGraphIs this causal?19
Roadmap1Data2Measuring Inflation across Income Groups3The Response of Supply to Market Size Effects1Effect of demand on supply2Do changes in the income distribution imply large inflation inequality?3Simple model20
Effect of Demand on SupplyGrowth of demand in a given part of product space over time dependson:IIIInitial spending shares of household groupsChanges in number of households in each groupChanges in per-capita spending of households groupsBartik-style research design[Bartik 1991; Blanchard and Katz 1992; Acemogluand Linn 2004; Dellavigna and Pollet 2007; Goldsmith-Pinkham, Sorkin and Swift 2016]:IIUse component of demand growth coming from change in number ofhouseholds, keeping spending share as in initial periodMeasure supply response using two outcomes: spending on newproducts and price changes for continued productsImplement using 108 age-income groups (9 income groups and 12 agegroups) and product-module-by-price-decile cells across product spaceIIChanges in age-by-income distribution measured in Current PopulationSurvey between 2000-2004 and 2011-2015Conduct analysis at national level21
Effect of Demand on SupplyGrowth of demand in a given part of product space over time dependson:IIIInitial spending shares of household groupsChanges in number of households in each groupChanges in per-capita spending of households groupsBartik-style research design[Bartik 1991; Blanchard and Katz 1992; Acemogluand Linn 2004; Dellavigna and Pollet 2007; Goldsmith-Pinkham, Sorkin and Swift 2016]:IIUse component of demand growth coming from change in number ofhouseholds, keeping spending share as in initial periodMeasure supply response using two outcomes: spending on newproducts and price changes for continued productsImplement using 108 age-income groups (9 income groups and 12 agegroups) and product-module-by-price-decile cells across product spaceIIChanges in age-by-income distribution measured in Current PopulationSurvey between 2000-2004 and 2011-2015Conduct analysis at national level21
Effect of Demand on SupplyGrowth of demand in a given part of product space over time dependson:IIIInitial spending shares of household groupsChanges in number of households in each groupChanges in per-capita spending of households groupsBartik-style research design[Bartik 1991; Blanchard and Katz 1992; Acemogluand Linn 2004; Dellavigna and Pollet 2007; Goldsmith-Pinkham, Sorkin and Swift 2016]:IIUse component of demand growth coming from change in number ofhouseholds, keeping spending share as in initial periodMeasure supply response using two outcomes: spending on newproducts and price changes for continued productsImplement using 108 age-income groups (9 income groups and 12 agegroups) and product-module-by-price-decile cells across product spaceIIChanges in age-by-income distribution measured in Current PopulationSurvey between 2000-2004 and 2011-2015Conduct analysis at national level21
Spending on Baby Diapers by Age GroupsFraction of Total Spending on DiapersAccounted For by Age Group (%)'403020100202530354045505560657075Household Age22
Spending Across Quality Ladder by Income GroupsShare of Spending (%)1312111098123456789Decile of Price per Ounce (within Product Module)Household Income 100k10Household Income 30k23
Annualized Growth Rate of Number of Households,2011-2015 relative to 2000-2004 (%)Changes in Income Distribution for 30-Year-Olds(CPS Data)43210050000100000150000200000Household Income (2004 )24
0.511.5Spending per Capita in 2013-2015 ( )2Relevance of Demand Growth Predictor0.511.522.5Spending per Capita in 2004-2006 ( )Coeff. 0.9114*** (s.e. 0.0301).Observation is household age-income group by product module by price decile.25
Results26
Effect of Demand on New ProductsAverage Share of Spendingon New Products, 2004-2015 (%)109.598.587.5.6.811.2Annualized Predicted Increase in Total Spending,2000-2004 to 2011-2015 (%)Coeff. 2.735*** (s.e. 0.488).26
Effect of Demand on Inflation for Continued ProductsAverage Annual Inflation Rate,2004-2015 (%)1.61.51.41.31.2.6.811.21.4Annualized Predicted Increase in Total Spending,2000-2004 to 2011-2015 (%)Nested CES Inflation Rate. Coeff. -0.435*** (s.e. 0.0907)27
Effect of Demand on Supply: Main ResultsShare of SpendingContinued Productson New Products (pp)Inflation Rate cted Increase in Spending,Annualized (%)Age and Income ControlsProduct Module Fixed EffectsYesYesR20.540.52Number of Observations10,75010,750Number of Clusters1,0751,075Standard errors clustered by product modulesInterpreting MagnitudesMore GraphsRobustness28
Roadmap1Data2Inflation across Income Groups3The Response of Supply to Market Size Effects1Effect of demand on supply2Do changes in the income distributionimply large inflation inequality?3Simple model29
Supply Response to Shifts in Income DistributionUse two ingredients to build inflation inequality implied by shifts inincome distribution:IHistorical changes in the income distribution to get changes in demand:dl snl · gnnwhere n denote 18 household income groups, with average growth rategn in 1996-2006 from CPS data GraphIPoint estimates to get new products and price changes on continuedproducts implied by change in demand:New ProductslImplied 2.73 · dlΠImplied 0.43 · dllCompare implied vs. actual relationships between new products/pricechanges and mean consumer income (Il n snl In ) across product spaceIResult: implied relationships account for 80% of actual relationships30
Supply Response to Shifts in Income DistributionUse two ingredients to build inflation inequality implied by shifts inincome distribution:IHistorical changes in the income distribution to get changes in demand:dl snl · gnnwhere n denote 18 household income groups, with average growth rategn in 1996-2006 from CPS data GraphIPoint estimates to get new products and price changes on continuedproducts implied by change in demand:New ProductslImplied 2.73 · dlΠImplied 0.43 · dllCompare implied vs. actual relationships between new products/pricechanges and mean consumer income (Il n snl In ) across product spaceIResult: implied relationships account for 80% of actual relationships30
Supply Response to Shifts in Income DistributionUse two ingredients to build inflation inequality implied by shifts inincome distribution:IHistorical changes in the income distribution to get changes in demand:dl snl · gnnwhere n denote 18 household income groups, with average growth rategn in 1996-2006 from CPS data GraphIPoint estimates to get new products and price changes on continuedproducts implied by change in demand:New ProductslImplied 2.73 · dlΠImplied 0.43 · dllCompare implied vs. actual relationships between new products/pricechanges and mean consumer income (Il n snl In ) across product spaceIResult: implied relationships account for 80% of actual relationships30
Average Share of Spendingon New Products, 2004-2015 (%)New Products From Shifts in Income roduct Modules by Price DecilesRanked by Mean Consumer Income (2006 )ActualPredictedOLS fit with actual outcome: Coeff. 1.2364*** (s.e. 0.1235).OLS fit with predicted outcome: Coeff. 1.0340*** (s.e. 0.00846).31
Average Annual Inflation Rate,Nested CES Price Index, 2004-2015 (%)Inflation Inequality From Shifts in Income 0Product Modules by Price DecilesRanked by Mean Consumer Income (2006 )ActualPredictedOLS fit with actual outcome: Coeff. -0.1912*** (s.e. 0.02886).OLS fit with predicted outcome: Coeff. -0.15938*** (s.e. 0.000682).32
Roadmap1Data2Inflation across Income Groups3The Response of Supply to Market Size Effects1Effect of demand on supply2Do changes in the income distributionimply large inflation inequality?3Simple model33
Overview of ModelGE model with free entry across sectors indexed by k andLit consumers of type i, with productivity Yi , in closed economyConsumers’ ProblemFirms’ ProblemClosed-Form SolutionsInterpreting MagnitudesKey ingredients: non-homothetic preferences and downward-slopinglong-term supply curve [Bresnahan and Reiss 1991; Acemoglu 1996, 2002,2007; Feenstra and Weinstein 2016; Comin, Lashkari, Mestieri 2016]Key prediction: given secular changes in the US income distribution,inflation inequality should be a long-term trend33
Conclusion34
Nested CES Inflation Rate, 2004-2015(Annualized, %)Lower Inflation for Higher-Income Households in Retail.21.81.61.41.2Below 30,000 30,000 to 100,000Household IncomeAbove 100,00034
. because Supply Responds to Shifting Demand .Average Annual Inflation Rate,2004-2015 (%)1.61.51.41.31.2.6.811.21.4Annualized Predicted Increase in Total Spending,2000-2004 to 2011-2015 (%)Nested CES Inflation Rate. Coeff. -0.435*** (s.e. 0.0907)35
Average Annual Inflation Rate,Nested CES Price Index, 2004-2015 (%). due to Changes in the Income 00Product Modules by Price DecilesRanked by Mean Consumer Income (2006 )ActualPredictedOLS fit with actual outcome: Coeff. -0.1912*** (s.e. 0.02886).OLS fit with predicted outcome: Coeff. -0.15938*** (s.e. 0.000682).36
Thanks!37
Approximation for New Products, Estimated Elasticities (Median of 6.5) Nested CES, Elasticity 11.5 from Broda and Weinstein (2010) Nested CES, Elasticity 7 from Montgomery and Rossi (1999) Nested CES, Elasticity 4 from Dube et al (2005) Nested CES, Elasticity 2.09 from Handbury (2013) Nested
May 02, 2018 · D. Program Evaluation ͟The organization has provided a description of the framework for how each program will be evaluated. The framework should include all the elements below: ͟The evaluation methods are cost-effective for the organization ͟Quantitative and qualitative data is being collected (at Basics tier, data collection must have begun)
Silat is a combative art of self-defense and survival rooted from Matay archipelago. It was traced at thé early of Langkasuka Kingdom (2nd century CE) till thé reign of Melaka (Malaysia) Sultanate era (13th century). Silat has now evolved to become part of social culture and tradition with thé appearance of a fine physical and spiritual .
Dr. Sunita Bharatwal** Dr. Pawan Garga*** Abstract Customer satisfaction is derived from thè functionalities and values, a product or Service can provide. The current study aims to segregate thè dimensions of ordine Service quality and gather insights on its impact on web shopping. The trends of purchases have
On an exceptional basis, Member States may request UNESCO to provide thé candidates with access to thé platform so they can complète thé form by themselves. Thèse requests must be addressed to esd rize unesco. or by 15 A ril 2021 UNESCO will provide thé nomineewith accessto thé platform via their émail address.
̶The leading indicator of employee engagement is based on the quality of the relationship between employee and supervisor Empower your managers! ̶Help them understand the impact on the organization ̶Share important changes, plan options, tasks, and deadlines ̶Provide key messages and talking points ̶Prepare them to answer employee questions
Chính Văn.- Còn đức Thế tôn thì tuệ giác cực kỳ trong sạch 8: hiện hành bất nhị 9, đạt đến vô tướng 10, đứng vào chỗ đứng của các đức Thế tôn 11, thể hiện tính bình đẳng của các Ngài, đến chỗ không còn chướng ngại 12, giáo pháp không thể khuynh đảo, tâm thức không bị cản trở, cái được
2. Realized Net Long-Term Gains and Corporate Equity of Households, 1954-1985 28 3. Ratio of Realized Long-Term Gains to Gross National Product, by Income Group, 1954-1985 30 4. Average Marginal Tax Rates on Long-Term Gains for Selected AGI Groups 39 5. Marginal Tax Rates on Long-Term Gains and the Ratio of Long-Term Gains to Gross National .
the elasticities is small: We estimate a short -term gains elasticity of -0.52 and a long -term gains elasticity of 1.0. We find that high income taxpayers are more respon sive with elasticities of -0.75 for short term gains and 1.5 for long term gains. We also find evidence that taxpayers minimize their tax liability by timing their gains and
