Business Incentives and Employment: What Incentives Work and Where

1
Business Incentives and Employment What
Incentives Work and Where?

• by
• William Hoyt
• Christopher Jepsen
• Kenneth R. Troske
• University of Kentucky

2
Introduction

• Offering incentives to businesses considering
  locating in a state, or offering incentives to
  try and keep businesses from leaving a state, has
  become wide spread in recent years.

3
Introduction

• One reason offering incentives remains
  controversial is the lack of rigorous empirical
  analysis assessing the effectiveness of these
  programs
• Researchers usually know the value of the
  incentive awarded to a firm but not the value
  of the incentive actually taken by the firm
• Since there are large differences between the
  amount of incentives awarded and the amount
  claimed only having data on incentives awarded
  will lead researchers to understate the
  effectiveness of incentives

4
Introduction

• We study incentives using data containing
  information on the incentives both awarded to and
  received by firms in Kentucky between 1992 to
  2004
• We evaluate the effect of these incentives on
  employment using panel data of Kentucky counties
• We focus on the effect of incentives on
  employment growth since job creation is a primary
  goal of the program

5
Introduction

• There are a couple of aspects about Kentucky that
  help in our evaluation
• Kentucky has a large number of counties (120), so
  we can study the impact of these programs on a
  relatively large number of small areas
• Also examine the impact of these programs on
  neighboring counties

6
Introduction

• There are a couple of aspects about Kentucky that
  help in our evaluation
• Kentucky shares a border with a number of
  different states, so we can examine whether these
  programs simply shift employment from one region
  of the state to another, or whether they help
  attract businesses from other states

7
Previous studies

• Somewhat limited literature examining the effect
  of economic development programs on economic
  growth
• Much of the previous research assumes that
  incentives are awarded to firms at randon, which
  we know is not true
• Previous researchers also tend to assume that the
  effect of incentives will be instantaneous

8
Previous studies

• Most informative and rigorous previous work on
  tax incentives is Greenstone and Moretti (2003)
• Compared changes in property values of counties
  that won million dollar plants with counties
  that lost
• Winning counties grow faster after winning
• They dont have data on size of the incentive so
  they cant say whether the benefits exceed the
  cost

9
Previous Studies

• Advantage of this study over most of the previous
  studies
• County-level information on actual incentives
  received by firms in a county as opposed to
  incentives awarded
• Data for a large number of years (1992-2005)
• Better controls for non-random receipt of
  incentives
• Allow the effect of incentives to occur with a lag

10
Data and Methodology

• Panel data of 120 Kentucky counties from 1992 to
  2005
• Preferred specification uses two-year lag for
  incentives, reducing panel to 1994 to 2005
• Data on incentives received in each county and
  year, from Cabinet for Economic Development
• Dependent variable is county-level employment

11
Kentuckys Business Incentive Programs

• For our study we divide the incentive programs
  into three groups
• Tax Incentives (KIDA, KREDA, KJDA, KIRA)
• Training Incentives (BSSC)
• Financing Incentives (KEDFA, EDB)

12
Data and Methodology

• Our analysis is based on regressions where
• The dependent variable is log employment in a
  county
• The key independent variables are the log amounts
  of tax, training and financing incentives
  received
• We also control for the demographic
  characteristics of county residents as well as
  other government expenditures and taxes and the
  year
• Also control for unobserved aspects of the county
  that are fixed over time

13
Data and Methodology

• When examining the impact of incentives on
  economic activity, we look at how incentives
  received two years ago impact economic activity
  in the current year
• For example, we look at the impact of incentives
  taken in 2000 on employment growth in a county in
  2002

14
Data and Methodology

• There are a number of reasons why it may take a
  couple of years before we see the economic impact
  of the incentives
• A firm may continue to grow for a number of years
  after receiving the incentive
• When a firm hires workers, these workers spend
  more money in a community, impacting other
  businesses in the area. It may take a couple of
  years for these indirect effects to be seen.

15
Data and Methodology

• In much of our work, we separate the counties
  into those that border other states and those in
  the interior of the state
• Counties on the border compete with other states
  (with different incentive programs) for companies
• Interior counties compete primarily with other
  Kentucky counties

16
Data and Methodology

• Other factors we consider
• Whether incentives have different effects in
  urban areas compared to rural areas
• Whether incentives have spillover effects into
  neighboring counties
• Using other measures of employment from BLS, we
  examine how incentives might impact employment
  for residents in a county

17
Results

• Start by looking at the use of business
  incentives over time and across different regions
  of the state

18
Amount of Business Incentives Taken as Percentage
of Earnings, 1992 to 2004
19
Amount of Business Incentives

• Overall amount of business incentives are fairly
  small relative to the size of the economy
• The combined amount of all programs in a year
  never exceeds 0.2 of total earnings in the state
• In 2003 the amount of the incentives was less
  than 10 of business taxes collected
• In 2003 amount of incentives was less than 1 of
  total state government revenue
• Incentives can represent a significant decrease
  in taxes for an individual firm

20
Tax Incentives Claimed as Percentage of Tax
Incentives Awarded
21
Value of Incentives Claimed vs. Awarded

• Only a small percentage of tax incentives awarded
  are ever claimed
• By 2004 only 12 of incentives awarded had been
  claimed
• Shows that it is important to use tax incentives
  claimed when studying the impact of incentives

22
Incentives as a Share of Earnings by Region
23
Ranking of Counties by Total Value of Incentives
Relative to Total Earnings
24
Effects of Business Incentives on Employment

• Results from our regressions show that
• Both training incentives and tax incentives have
  positive and significant effects on employment in
  a county, but only in border counties they have
  no significant effect in interior counties
• Financing incentives never have significant
  effects on employment in any county

25
Effects of Business Incentives on Employment
26
Size of the Effects

• Using these coefficients, we can estimate the
  impact of a 10 increase in incentives received
  by businesses operating in a county on employment
  in a county
• Mean training incentive in border counties is
  33,927, so a 10 increase is 3,393
• Mean tax incentives in border counties is
  551,204, so a 10 increase is 55,120
• Mean employment in border county is 26,986

27
Effects of Business Incentives on Employment
28
Effects of Business Incentives on Employment

• Employment effects seem small, but keep in mind
• The spending on these programs is small relative
  to the size of the economy
• Training incentives average about 0.01 of
  earnings while tax incentives average about 0.08
  of earnings
• Effects are large relative to other estimates of
  the employment effects of a tax cut

29
Effects of Business Incentives on Employment

• Several possible reasons why training incentives
  have large effects relative to their size
• Amount available was restricted by the
  legislature (2.5 million/year)
• Low skill of workers is a serious problem in
  Kentucky
• Training provides workers with skills which they
  keep even if the are no longer working for the
  firm

30
Differing Effects of Incentives Between non-MSA
and MSA Border Counties
31
Differing Effects of Incentives Between non-MSA
and MSA Border Counties

• Training incentives have a positive and
  significant impact on employment in both MSA and
  non-MSA border counties
• Tax incentives have a positive and significant
  impact on employment only in MSA border counties

32
Spillover Effects of Incentives

• We examine the impact that incentives received by
  a business operating in one county has on
  employment in surrounding counties
• We find no significant effect on employment in
  surrounding counties
• Incentives do not appear to simply shift
  employment from one county to another within a
  state, nor do they lead to growth in economic
  activity in other counties

33
Long-Run Effects of Incentives

• We find that the effects of incentives on county
  employment are persistent
• We find significant effects for training
  incentives for three years after the incentives
  were received
• We find significant effects of tax incentives for
  five years after the incentives were received

34
Effects of Incentives on Unemployment and Labor
Force Participation

• We looked at how incentives lead to an increase
  in employment
• We find that training incentives tend to increase
  the size of the labor force in a county
• These incentives work by drawing workers back
  into the labor market
• We find that tax incentives work by reducing the
  number of unemployed workers in a county

35
Summary

• Training and tax incentives have positive effects
  on employment in border counties
• Training incentives have positive effects in all
  border counties while tax incentives only have
  significant effects in MSA border counties
• Financing incentives have no affect in any county
• No evidence of spillover effects between counties
• Effects appear to persist for several years

36
Concluding Thoughts

• I find the results for training incentives one of
  the more interesting findings in the study
• The effects are quite large given that relatively
  little money is spent on the program
• This is the only program to have any impact in
  rural areas
• We would like to learn more about how this
  program is administered and what training is
  provided
• Training appears to be fairly general

37
Concluding Thoughts

• It would be interesting to compare the effect of
  receiving training through ones firm with the
  effect of training received through federal
  programs such as the Workforce Investment Act
  (WIA)

38
Thank you