The Economic Multiplier Impacts of Biopharmaceutical Related Industries in Massachusetts on Outputs, Earnings, and Employment in the Overall State Economy

The Economic Multiplier Impacts of
Biopharmaceutical Related Industries in Massachusetts
on Outputs, Earnings, and Employment in the
Overall State Economy

PhRMA Research Paper No. 14



Prepared by:
Andrew Sum
Ishwar Khatiwada
Joseph McLaughlin
Paulo Tobar
With
Sheila Palma

Center for Labor Market Studies
Northeastern University
Boston, Massachusetts

Prepared for:
The Pharmaceutical Research and Manufacturers of America
(PhRMA)



August 2007

---

Table of Contents


Introduction..................................................................................................................................... 2

Input-Output Models and the RIMS Models of the U.S. Bureau of Economic Analysis............... 3

Alternative Types of Economic Multipliers in Input-Output Analysis .......................................... 5

Appendix A: Estimates of Output, Earnings, and Employment Multipliers for
Biopharmaceutical-Related Industries in the U.S., Massachusetts, and All Other States
Across the Country

Appendix B: Estimates of the Number and Industrial Distribution of the Jobs Created by An
Increase of $1 Million in Final Sales by Biopharmaceutical-Related Industries in
Massachusetts and the U.S.


1

---

Introduction
The direct effects of biopharmaceutical-related industries in Massachusetts on a variety
of economic and labor market outcomes were examined in a number of previous research papers
in this series.1 These research papers described and assessed the outputs, research and
development expenditures, financial performance, and employment levels of these industries
both at various points in time and over time. The demographic and socioeconomic traits of
workers in these industries, the occupational characteristics of their jobs, and their annual
earnings also were analyzed.
The economic and labor market impacts of an industry often go well beyond their direct
effects on output, earnings, and employment in the regional, state or local economy in which
they are located. Industries, especially those that export their goods or services outside of the
state in which they are located, can generate important multiplier effects on the rest of the state
economy.2 These multiplier effects can include both indirect effects on other state industries
through the purchases of inputs by an exporting firm from other state firms and the induced
effects on other industries from the local consumption spending of workers and owners in the
exporting industries.3 Knowledge of the magnitude of these multiplier effects of individual
industries is indispensable for evaluating their overall contributions to the state economy.
As we have shown in an earlier research paper, a major share of the output of
biopharmaceutical related industries in Massachusetts is exported outside of the state, including
sales to other countries.4 Biopharmaceutical industries, thus, constitute part of the export base of
Massachusetts and can generate multiplier effects on the output and employment of industries in

1 See: (i) Joseph McLaughlin, Ishwar Khatiwada, and Andrew Sum, Recent Trends in Payroll Employment in
Massachusetts in the Pharmaceutical and Medicine Manufacturing, Surgical and Medical Instrument Manufacturing
Industries, and Physical, Engineering, and Biological Research Industries of Massachusetts: A Comparative
Perspective, PhRMA Research Paper #2, December 2006; (ii) Ishwar Khatiwada and Andrew Sum, Annual Earnings
Levels and Trends in the Real Annual Earnings of Workers in Biopharmaceutical-Related Industries in
Massachusetts and the U.S. PhRMA Research Paper #7, January 2007; and (iii) Ishwar Khatiwada and Andrew
Sum, The Financial Stock Market, and R&D Expenditures Performance of PhRMA Member Companies Operating
in Massachusetts in Recent Years, PhRMA Research Paper #9, April 2007.
2 In some cases, the exports of services and goods involve customers coming into the state from other states or
countries to make these purchases, such as tourists or college students coming into Massachusetts colleges and
universities from other states.
3 Industries that produce goods and services for local sale that can substitute for imports from other states and
countries also can generate multiplier effects on the state economy.
4 Research and development activities financed by public and private monies from outside the state also can be
treated as export-oriented activities.

2

---

the rest of the economy. Due to their high levels of purchases from other industries in the state
and the high levels of induced local spending by workers and managers in these industries,
biopharmaceutical related industries in Massachusetts and the U.S. tend to generate relatively
high output, earnings, and especially employment multipliers. In this research paper, we will
define, describe, and analyze a variety of output, earnings, and employment multipliers for four
biopharmaceutical-related industries in the state and compare findings for Massachusetts with
those for the nation and all other states. The values of these multipliers were provided to the
Center for Labor Market Studies by the U.S. Department of Commerce’s Bureau of Economic
Analysis under its Regional Input-Output Modeling System more widely known by its acronym
(RIMSII).5

Input-Output Models and the RIMS Models of the U.S. Bureau of Economic
Analysis
The economic multipliers for state biopharmaceutical-related industries that will be
presented in this paper are based on a comprehensive set of national/state input-output models
developed by the U.S. Department of Commerce. An input-output model provides a statistical
depiction of the inter-industry linkages within a national or state economy.6 The national or state
economy is divided into a set of industries. For each industry, its gross output is allocated to
either final sales (households, capital goods sales to industries, government, exports to other
states/nations).7 These linkages in terms of sales to other industries are referred to as forward
linkages. The input-output models also trace purchases of each state industry from other
industries in the state. These linkages are referred to as backward linkages.
The data on the gross outputs of industries and these inter-industry linkages are used to
calculate a matrix of input-output coefficients (or technical coefficients). This set of data on

5 See: U.S. Department of Commerce, Bureau of Economic Analysis, Regional Multipliers: A User Handbook for
the Regional Input-Output Modeling System (RIMSII), (Third Edition, 1997), U.S. Government Printing Office,
Washington, D.C., 1997.
6 For an introduction to input-output analysis and its applications to state and regional economies,
See: (i) H.B. Chenery and P.J. Clark, Interindustry Economics, John Wiley, New York, 1962; (ii) M. Javin Emerson
and F. Charles Lamphere, Urban and Regional Economics: Structure and Change, Allyn and Bacon, Inc., Boston,
1975; (iii) William H. Miernyk, The Elements of Input-Output Analysis, New York, Random House, 1965; (iv)
William H. Miernyk, et al., Impact of the Space Program on a Local Economy: An Input-Output Analysis, West
Virginia University, Morgantown, 1967.
7 Annual gross output is the value of all revenues received from sales of the product or services plus the change in
inventories. Output produced but not sold is part of the gross output of the industry.

3

---

gross outputs, final demand sales, and technical coefficients is used with a set of matrix algebra
calculations to calculate the direct and indirect output requirements from each industry of the
state to produce a dollar of final sales from each industry. By incorporating the economic links
between households and each industry into the input-output model, we can also calculate the
induced effects of an expansion of final sales from each industry. 8 As will be described more
fully below, these induced effects are brought about by the spending of workers, managers, and
owners of the industry on the goods and services produced by other industries across the state.
During the 1970s, the U.S. Department of Commerce’s Bureau of Economic Analysis
(BEA) developed a statistical methodology to estimate regional Input-Output (I-O) multipliers
known as the Regional Industrial Multiplier System (RIMS). In the 1980s, the BEA enhanced
this methodology, and it was called the Regional Input-Output Modeling System II (RIMS II).
The multipliers for industries are based on complex set of Input-Output (I-O) tables originally
developed by Wassily Leontief for which he was awarded the Nobel Prize In Economics in 1973.
The I-O table shows for each industry the value of the inputs it purchased from other industries
and outputs sold to other industries. RIMS II multipliers are available for 473 industries or 60
industry aggregates in the I-O table.9 The Bureau of Economic Analysis produces these
multipliers for the nation, geographic regions (New England, Mid-Atlantic), individual states,
and major metropolitan areas. In this paper, we primarily rely on BEA multipliers at the state
level for four major biopharmaceutical related industries.
There have been numerous input-output studies conducted in the U.S. on various
industries to assess their economic impacts on regions, states, or the nation as a whole. For
example, a recent study on the contribution of the biofuels industry to the Iowa economy used a
variety of multipliers generated by BEA.10 There also are studies on the economic impacts of
hospitals, colleges, drug industries, the aviation industry, tourism, transportation, healthcare,
biotechnology, and venture capital, at the local, state, and national level.11 These studies use

8 Households both provide labor and capital services (bonds, stocks, land) to an industry for which they receive a
stream of income and they purchase goods and services from industries.
9 See: U.S. Department of Commerce, Bureau of Economic Analysis, Regional Multipliers: A User Handbook for
the Regional Input-Output Modeling System, (RIMS II), U.S. Government Printing Office, Washington, D.C., 1997.
10 See: John M. Ubarchuk, “Contribution of the Biofuels Industry to the Economy of Iowa,” Prepared by LEGC
LLC, Prepared for the Iowa Renewable Fuels Association, February 2007.
11 See: (i) “The Economic Impact of the Biotechnology and Pharmaceutical Industries in New York,” Published by
Office of the State Comptroller, New York, February 2005; (ii) Ernst and Young, “The Economic Contribution of
the Biotechnology Industry to the U.S. Economy,” Prepared for the Biotechnology Organization, February 2006;

4

---

multipliers from the BEA as well as those based on their own simulations that rely on
sophisticated computer modeling.

Alternative Types of Economic Multipliers in Input-Output Analysis
There are a number of different types of multipliers that can be generated by input-output
models. The first general category of multipliers consists of Type I and Type II multipliers. The
Type I multiplier captures only the direct and indirect effects of a change in final demand sales,
earnings, or employment in a given industry. For example, a Type I final demand output
multiplier would capture the direct effect of an expansion in final demand sales of industry i plus
all the indirect effects via the changes in the outputs of all other industries in the state brought
about by industry i’s purchases from other industries. As a pharmaceutical manufacturing
industry increases its sales to consumers in the state or to consumers in other states and
countries,12 the industry will directly purchase inputs of goods and services from other state
industries, including itself. These industries in turn will buy goods and services from other state
industries to produce additional inputs, which will then initiate another cycle of inter-industry
spending of increasingly smaller sizes.13 The combined effects of these second and third round
and additional rounds of spending are the indirect effects of the increase in final demand sales of
industry i. The higher the purchases by industry i from other state industries (i.e., these inter-
industry linkages), the higher will be the size of these indirect effects. The value of the Type I

(iii) The Campbell-Hill Aviation Group, Inc., “Commercial Aviation and The American Economy”, Prepared for the
Air Transport Association of America, March 2006; (iv) Global Insight Inc., “The Economic Importance of Venture
Capital Backed Companies to the U.S. Economy”, Prepared for National Venture Capital Association, 2007; (v)
Eastern Research Group Inc., “Economic Analysis for Final Effluent Limitations Guidelines and Standards for the
Pharmaceutical Industry”, Report Prepared for U.S. Environmental Protection Agency (EPA), July 1998; (vi)
American Hospitals Associations, “Beyond Healthcare: The Economic Contributions of Hospitals”, 2006; (vii)
“Economic Impact of Skidmore College on Saratoga County”, Prepared by Capital District Regional Planning
Commission, Albany, New York, April 2007; (viii). Ross DeVol, Rob Koepp with Perry Wong and Armen
Bedroussian, “The Economic Contributions of Health care to New England”, A Report Prepared by Milken Institute
for New England Healthcare Institute, February 3003; (ix) The Boston Consulting Group, “MassBiotech 2010:
Achieving Global Leadership in The Life-Sciences Economy”, Prepared for Massachusetts Biotechnology Council,
2002; (x) Tim Lynch, “Florida High Speed Ground Transportation Economic Benefit and Cost Impact Restudy and
Public Transportation Financing and Subsidies by Mode in the United States”, Econometrics Consultant Inc.,
August 2002.
12 In a state input-output model, sales to consumers within the state are treated as consumption expenditures under
the final demand matrix while sales to consumers in other states and countries would be classified as export sales
under the final demand matrix.
13 In input-output models, the size of the direct and indirect requirements from each industry to produce a dollar of
gross output from industry i are estimated by inverting the [I – A] matrix of coefficients where A is the technical
coefficients matrix and I is the identify matrix.

5

---

multiplier for a given industry is calculated by dividing the sum of the direct and indirect effects
by the direct effects (Table 1).
Table 1:
The Formulas for the Two Types of Multipliers in Input-Output Analysis

Type I Multiplier
=
Direct + Indirect Effects
Direct Effects

Type II Multiplier
=
Direct + Indirect + Induced Effects
Direct Effects

The Type II multiplier captures the direct, indirect, and induced effects of an expansion in
final sales, earnings, or employment in a given industry. The induced effects capture the impacts
of the spending of workers, managers, and owners in the affected industry. When an industry
expands its output level, it will create a stream of wages and salaries for its workforce and a
stream of profits/interest/rents for its owners. These workers and owners will spend a portion of
their incomes on goods and services and new housing produced by other state industries, thereby
expanding output, earnings, and employment in these other state industries. The higher the wages
and salaries of workers in the industry, the higher their degree of residency in the state, the
higher their propensity to consume locally produced goods and services, and the higher the
geographic concentration of owners in the state, the higher will be the size of these induced
effects. As revealed in several of our earlier research papers, mean annual wages and salaries of
biopharmaceutical industry workers in Massachusetts are considerably above the statewide
average, and the vast majority of workers in these industries reside in the state.14 Both of these
developments will raise the induced effects from an expansion of final demand sales and
employment in biopharmaceutical industries, thereby increasing the value of Type II multipliers.
The value of the Type II multiplier for a given industry is calculated by dividing the sum of the
direct, indirect, and induced effects by the direct effect (Table 1). The higher the indirect and
induced effects, the higher will be the value of the Type II multiplier.

14 See: Ishwar Khatiwada and Andrew Sum, Annual Earnings Levels and Trends in the Real Annual Earnings of
Workers in Biopharmaceutical-Related Industries in Massachusetts and the U.S. PhRMA Research Paper #7,
January 2007; and Andrew Sum, Ishwar Khatiwada with Meredith Franks and Sheila Palma, The Annual Earnings
of Workers in Biopharmaceutical-Related Industries of Massachusetts and the U.S.: A Comparative Assessment,
Research Paper #10, April 2007.

6

---

A variety of economic multipliers can be produced through the application of input
output analysis particularly when the I/O analysis is supplemented by earnings and employment
data by industry. Among the types of multipliers used in economic impact studies at the national,
state, and local level are output, earnings, income, and employment multipliers. In its RIMS
modeling, the U.S. Bureau of Economic Analysis produces values for the following five types of
multipliers (See Table 2 for a description of the formula for each type of multiplier).
Final demand output multipliers
Final demand earnings multipliers
Final demand employment multipliers
Direct-effect earnings multipliers
Direct-effect employment multipliers

Table 2:
Algebraic Formulas for the Five Types of Industry Multipliers
Available from the BEA RIMS Modeling at the State Level

(1) Final demand output
=
∆ in Gross Output of All State Industries
multiplier of industry i
∆ in Final Demand of Industry i
(2) Final demand earnings
=
∆ in Earnings of All Workers in State
multiplier of industry i
∆ in Final Demand of Industry i
(3) Final
demand
=
∆ Total Employment in State
employment multiplier of
∆ in $1 million dollars in final demand sales of
industry i
industry i
(4) Direct earnings multiplier =
∆ in Earnings of All Workers in State
of industry i
∆ in Earnings of Industry i
(5) Direct
employment
=
∆ in Total Employment in State
multiplier of industry i
∆ in Employment in Industry i

Each of the final demand multipliers is designed to measure the change in the outcome
being analyzed (output, earnings) from a given level of change in final demand delivered by the
industry being analyzed. For example, the final demand output multiplier represents the change
in the gross output of all industries in a state from a change in one-dollar of final demand sales
from a given industry (e.g., pharmaceutical manufacturing). The output multiplier captures
direct, indirect, and induced output effects of the increase in final demand.

7

---

The final demand earnings multiplier captures the increase in the total earnings from
employment of all household members in the state from a change in final demand sales from
industry i. The value of this multiplier will be influenced by the size of the direct, indirect, and
induced earnings effects from an expansion in final demand sales of the industry. The size of the
direct earnings effects of biopharmaceutical industries vary across industry subgroup. They are
smaller in more capital intensive and R&D intensive industries such as pharmaceutical and
medicine manufacturing where the earnings of workers are a smaller share of gross output. The
direct earnings effects are large in more human capital intensive industries such as scientific
research and development industries where the earnings of workers employed by the industry
account for a higher share of the gross output of the industry.
The third multiplier is the final demand employment multiplier of industry i. This
multiplier measures the number of additional jobs that will be created across all industries of the
state from a $1 million increase in final sales of industry i. A final demand employment
multiplier for an industry equal to 8.2 implies that an additional million dollars in final demand
sales will create 8.2 jobs in the state’s economy. These job creation effects capture the direct,
indirect, and induced employment effects of this additional billion dollars in final sales.15 For this
employment multiplier, the BEA also identifies the major industrial sectors in which these
additional jobs will be created by an expansion of final sales in industry i.
There are two other types of multipliers produced by the U.S. Bureau of Economic
Analysis for industries in each state. They are the direct earnings multiplier and the direct
employment multiplier. The definitions of these two alternative multipliers are displayed below.
The direct earnings multiplier. This multiplier represents the change in the earnings of
all households in the state from a one-dollar change in the earnings of workers in the state
industry.
The direct employment multiplier. This multiplier represents the change in total
employment in the state from an increase of one job in the specific state industry being
examined. An employment multiplier of 5.0 for a given industry would imply that an increase of

15 The key assumption underlying these calculations is that the marginal change in employment per dollar of output
in each industry is equal to the average ratio of employment per dollar of gross output in each industry. For this
relationship to hold true, there should be no significant degree of underemployment of workers in these industries at
the time of the expansion in sales.

8

---

one job in the industry under analysis (i.e., pharmaceutical manufacturing) would directly and
indirectly create five jobs in the state economy.
The U.S. Bureau of Economic Analysis provides multiplier estimates for specific NAICS
industries at the national, regional, and state level. The national multipliers for a given industry
will always be greater than the value of the state multipliers for a given industry for several
reasons. First, the indirect effects of an industry’s expansion will be greater at the national level
than at the state level since some of the purchases of goods and services by the industry will take
place outside of the state. Interindustry purchases outside of the state but inside the U.S. will add
to the national multiplier but not the state multiplier for that industry. Second, the induced effects
of an industry’s expansion will be greater at the national level than at the state level since some
of the purchases of the industry’s workers and owners will take place outside of the state rather
than inside the state. This will be especially true when the workers and owners of the industry do
not live in the same state in which the industry is located. Their induced spending will create
incomes and jobs in surrounding states or states in other parts of the country.
The U.S. Bureau of Economic Analysis provided us with values of national and state
multipliers for five biopharmaceutical-related industries. We selected four of these industries for
our multiplier analysis. The NAICS industry codes and titles of these four industries are
presented below:
NAICS 3254000, Pharmaceutical and Medicine Manufacturing
NAICS 339112, Surgical and Medical Instrument Manufacturing
NAICS 339113, Surgical Appliance and Supplies Manufacturing
NAICS 571700, Scientific Research and Development Services
For each of these four industries, we obtained estimates of four multipliers: final demand
output multipliers, final demand employment multipliers, direct effect earnings multipliers, and
direct-effect employment multipliers for the U.S., Massachusetts, and each of the other 49 states
and the District of Columbia. We calculated Massachusetts’ ranking among the 50 states and DC
for each of these three multipliers for each of the four industries. Findings are displayed in
Tables 3 to 6.

9

---