Information about http://inforumweb.umd.edu/papers/inforum/info/Inforum_Overview07.pdf
Department of Economics …
Tags: annual conferences, clopper almon, crucible, department of economics, econometric models, economic education, economic models, economic projections, economic research, economics department, economics university, empirical economics, fertile ground, ierf, inforum partners, macroeconomic models, private sector organizations, professor emeritus, trujillo spain, werling,Pages: 10
Language: english
Created: Fri Jul 20 10:58:31 2007
Display cached document
Page 1
Page 2
Page 3
Page 4
Page 5
Page 6
Page 7
Page 8
Page 9
Page 10
Department of Economics
University of Maryland
http://inforum.umd.edu
Contact:
Jeff Werling
(301) 405-4607
werling@econ.umd.edu
June 2007
Inforum Spring 2007
Introducing Inforum
Inforum is the popular name for a not-for-profit economic education and research corporation, the
Interindustry Economic Research Foundation (IERF). Since its founding over thirty five years ago,
Inforum 1 has been dedicated to improving business planning, government policy analysis, and the
general understanding of the economic environment. It accomplishes this mission through:
· Building and using structural economic models of U.S. and other economies. Inforum pioneered
the construction of dynamic, interindustry, macroeconomic models which portray the economy
in a unique "bottom-up" fashion.
· Working with government and private sector organizations to investigate a variety of issues.
Economic projections and analysis using Inforum econometric models are distinguished by
detail at the industrial and product level.
· Serving as a training crucible for University of Maryland graduate students. Students receive
valuable training in empirical economics and find fertile ground for dissertation research.
· Maintaining active ties with a world-wide network of research associates, each of which uses
Inforum modeling methods and software. The Inforum partners have held annual conferences
since 1993. The 2007 conference will be held in Trujillo, Spain in September.
Inforum's operations are housed at the University of Maryland in the Economics Department. It was
founded there by Dr. Clopper Almon in 1967, now Professor Emeritus of the University. Clopper
supervised over 40 Ph.D. dissertations, many of which have contributed to the infrastructure of Inforum.
Most of our analyses involve the development and use of Interindustry-Macroeconomic (IM) models
that combine input-output structure with econometric equations in a dynamic and detailed framework.
Because of their ability to portray the detailed structure of economies over actual time periods, we find
that these models fill an important gap in the inventory of existing models of the U.S. and foreign
economies. For example, we often use these models to answer "what if" questions on the impact across
industries of fluctuation in the macroeconomic environment, such as changes in exchange rate or tax
policy. The effects on demands, revenue, production and trade can be described at a level of 97 sectors,
or for many scenarios, at the 360 sector-level.
Inforum services include macroeconomic and industrial forecasting and "satellite" modeling to connect
data for more detailed sectors to a more aggregated environment. Indeed, many subscribers use our
software and models on their computers for routine analysis or issue-specific research. We also perform
analytical research and computations of economic or other data, with particular expertise in input-output
techniques, global economic data, and international market comparisons.
Inforum researchers learn the details concerning the meaning and compilation of economic data so that
this data can be deployed in the most relevant and meaningful fashion. We are dedicated to timely,
thorough and reliable assistance to our research sponsors. Finally, Inforum explores economic
phenomena and principles in a nonpartisan fashion, according to generally accepted economic theory
and econometric methods, regardless of the implications for public policy or private strategy.
1
Inforum stands for the INterindustry FORecasting at the University of Maryland and is a registered trademark of
IERF. IERF handles contracts and subscriptions; a substantial portion of its receipts are donated to the University of
Maryland where the research is accomplished. Please visit our website at: www.inforum.umd.edu.
Inforum Spring 2007
Inforum Models
Most of Inforum's analyses are based on Interindustry-Macroeconomic (IM) models that combine input-
output structure with econometric equations in a dynamic and detailed framework. Inforum's flagship
model, Lift (Long-term Interindustry Forecasting Tool), is a 97-sector general equilibrium
representation of the U.S. economy that employs a "bottom-up" approach to macroeconomic modeling.
In an IM model, quantities such as total capital investment, total imports and total profit income are not
projected directly but are computed from the sum of their parts: investment by production branch,
imports by commodity, and profits by industry. This bottom-up technique possesses several desirable
properties for analyzing the economy. For example, industry-specific changes in government policies or
market setting can be individually specified, and the impact of these industry-level changes on related
sectors and on the aggregate economy can be identified. At the same time, the impact of
macroeconomic events such as exchange rate changes or fiscal policy can be traced to and illustrated at
the sectoral level.
Lift's structure contains information and simulation results for macroeconomic sectors and variables
such as the GDP and its components, inflation, employment, financial markets, government accounts,
and international balances. At the sectoral level, the model provides data and results for revenue and
production, consumption and government demand, capital investment and profits, employment and
wages, and exports and imports.
Inforum builds and maintains other models of the U.S. economy including:
· Iliad, a 360 sector model that uses the results of Lift to develop forecasts and simulations at a more
detailed level.
· Demographic Projection Model (DPM), a population forecasting system that produces forecasts by
gender and age.
· State Employment Modeling System (STEMS), a model which allocates national Lift results for
industrial production, employment and income by each state.
In addition, Inforum's International System of Models links the U.S. economy with its 12 major trading
partners. Specifically, in cooperation with research partners throughout the world, Inforum maintains
and uses Interindustry Macro Models (i.e., similar to Lift) for several countries including:
Austria Italy
Belgium Japan
Canada Mexico
China Spain
France South Korea
Germany United Kingdom
These models are tied together at the detailed commodity level through a Bilateral Trade Model (BTM).
This model has the capability to perform unique studies on global developments such as changing
exchange rates or new trade liberalization.
Inforum Spring 2007
Inforum Client and Project Examples
Federal Agencies:
· Center for Medicare and Medicaid Services(CMS)
Illustrate the long-term implications of increases in social and private medical expenditures.
Develop Demographic Projections Model (DPM).
· Department of Defense
Develop and maintain DEPPS, the Defense Employment and Purchases Projection System,
which calculates the macroeconomic, industrial, and regional effects of defense spending.
· Department of Commerce
Describe the economic impacts of energy price and exchange rate fluctuations.
· Institute for Defense Analysis
Analyze effects of defense force mobilization.
· Congressional Budget Office
Analyze budgetary impacts of defense expenditures.
· Other U.S. Government
Build and maintain interindustry-macroeconomic models for several foreign countries and an
international bilateral trade model. Analyze global bilateral trade flows and issues.
Private Sector:
· Energy Security Leadership Council /Securing America's Future Energy
Evaluate economic impacts of Senate energy bills.
· Axiom Valuation
Provide very detailed industry forecasts for online business web site.
· National Rural Electric Co-op Association
Develop county/utility area model to analyze electricity policy and regulation changes.
· Integra Information, Inc.
Develop and maintain detailed sales forecasts for U.S industries.
· Manufacturer's Alliance (MAPI)
Conduct special studies on the Asian Crisis and foreign trade liberalization.
· American Council of Life Insurance
Provide U.S. Demographic Projections Model (DPM) forecasts and analysis.
· PWC Consulting
Analyze effects of changing U.S. tax policies.
Foreign Clients:
· Canadian Government · Ministry for International Trade and
· Fundación Tomillo (Spain) Industry (Japan)
· Institute for International Trade and · Korean Environmental Institute
Investment (Japan)
Inforum Spring 2007
Inforum Staff
Clopper Almon, Jr.
Clopper founded the INFORUM project in 1967 and now is Professor Emeritus in the Economics
Department at the University of Maryland. He remains the Chairman of the IERF Board. He wrote The
Craft of Economic Modeling which he used to teach model building through hands-on experience with his
econometric software package G (available free at the Inforum web site). He created Interdyme, the
software environment used for developing Inforum's multisectoral policy and forecasting models.
Margaret McCarthy
Margaret works on the development and use of US databases and models. Margaret consults on U.S.
model-related projects, recently supplying, for example, an updated IO table economic time series data for
the U.S. economy to MITI. She leads Inforum's assistance to the Chief Actuary of the Centers for
Medicare and Medicaid. Margaret has 30 years of experience at Inforum, concentrating on the U.S. input-
output accounts and related U.S. industry statistics, as well as developing the family of Inforum models.
Douglas Meade
Doug returned to Inforum in May, 2006, after serving 3 years as Deputy Chief of the Industry Division at
the Bureau of Economic Analysis (BEA). Before working for BEA, Doug contributed significantly to the
development of the Lift and Iliad models of the U.S., and the Jidea model for Japan. He has served as
principal investigator on a wide variety of projects for private sector and government clients. He has
nearly 20 years of experience with economic modeling and data development, and has also held positions
at Data Resources Inc., and the Bureau of Census. Doug received his Ph.D. from the University of
Maryland.
Douglas Nyhus
Doug develops and maintains the international interindustry models and the bilateral trade system linking
countries into a world forecasting system. He performs many analyses with the international/bilateral
trade system. He has supported impact analyses by U.S. government agencies for studies of the Chinese,
Korean, Japanese, and Indian economies. He has nearly 30 years of experience working with international
models and data, and has been instrumental in developing the Inforum international network of
researchers. Doug has a PhD from the University of Maryland.
Jeffrey Werling
Returning to Inforum in 2003, Jeff has served as manager and principle investigator for several recent
projects, including a recent currency study for the Department of Commerce using the Inforum Bilateral
Trade Model. He maintains and operates the DEPPS model for the Department of Defense. Previously,
Jeff held positions as an international and industry economist with the National Electrical Manufacturers
Association (NEMA), the Manufacturers Alliance (MAPI), and the WEFA Group (now Global Insight).
Jeff received his PhD from the University of Maryland.
Ron Horst
Ron's work has included the estimation of investment, consumption and labor productivity equations for
the Lift model and maintaining Inforum software. He constructed a regional model of Andalusia, Spain.
He is completing his dissertation concerning at the economics of the nuclear power industry.
Inforum Spring 2007
The Inforum International System of Models
In order to analyze the global economic impact of various shocks (economic and non-economic), one
must use a consistent and empirical economic framework rooted in sound economic theory and reliable
economic data. Inforum pioneered such an infrastructure through its International System (IS) of IM
models of key national economies.
This collection of models has been built over the past three decades by a network of international
researchers applying a framework and philosophy first pioneered by Professor Clopper Almon at the
University of Maryland's Inforum project. The system currently includes models of Austria, Belgium,
Canada, China, France, Germany, Italy, Japan, Mexico, South Korea, Spain, the United Kingdom, and
the United States. The models are tied together through a Bilateral Trade Model (BTM) that describes
bilateral merchandise trade flows among countries for 120 commodities.
The Inforum system incorporates several important features that make it an ideal tool to provide
quantitative analyses on how the changing global environment is impacting not only individual
economies, but specific industries as well. These features include:
· National economies are modeled from the bottom-up.
While each model has an explicit accounting for macroeconomic quantities such as GDP, inflation
and unemployment, the approach to determining these quantities is unique. Combining a classical
input-output formulation with extensive use of regression analysis, the models employ a "bottom-
up" approach to macroeconomic modeling. For example, aggregate consumption, total exports, and
employment are not determined directly, but are computed as the sum of their parts: consumption of
specific goods and services, exports by commodity, and employment by industry.
· The national models reflect specific country structures and institutions.
Though the Inforum models share a common structure, they are not simple applications of a general
model form. At the most fundamental level, each model is based on the economic data of its subject
country, including the interindustry structure indicated by country-specific input-output tables.
Moreover, the models incorporate economic characteristics and econometrically-estimated behavior
specific to each country. The model of China reflects the institutions that are in place today and
projects how those institutions might change over time. Likewise, the US model assumes US
institutions and economic structures. Thus, each model will react to global economic shocks in
different ways.
· Scenario modeling capability is extensive and flexible.
The Inforum models possess a flexible structure for performing alternative scenarios. Exogenous
assumptions can be easily modified, and endogenous variables or equation structure can be altered at
the industry-level. For example, the system was successfully used to conduct an ambitious analysis
of European Integration that required more than a dozen sector-specific studies. Issues analyzed
included the removal of customs stations, changes in the structure of retail banking, the decline of
costs in trucking, and the reduction of management "x-inefficiencies" due to the increase of
competition in key sectors.
· The national models are linked via a bilateral trade model.
The Inforum models are connected by the detailed Bilateral Trade Model (BTM). As the name
implies, BTM models the bilateral flows of merchandise trade from one country to another for 120
categories of tradable goods. For each country and commodity, the model predicts the share of
Inforum Spring 2007
imports sourced in each of the other 13 countries plus two regions (Rest of Europe and Rest of
World). These shares are projected as a function of the relative prices among competing exporters
and each exporter's capacity to export. Thus, for example, an increase in the capital stock of the
automobile industry in South Korea will lead, over time, to an increase Korea's share in the imports
of other countries' auto imports. A detailed description of the BTM can be found at:
www.inforum.umd.edu/WorkPaper/INFORUM/Dissertations/Wang.pdf
Bilateral Trade Model Commodity Titles
1 Unmilled cereals 41 Other wood products 81 Constr,ming,oil equip
2 Fresh fruits,vegetables 42 Furnitures and fixtures 82 Metal, wood machinery
3 Other crops 43 Pulp and waste paper 83 Sewing & knit machines
4 Livestock 44 Newsprint 84 Textile machinery
5 Silk 45 Paper products 85 Paper mill machines
6 Cotton 46 Printing,publishing 86 Printing machines
7 Wool 47 Basic chemicals 87 Food-process machines
8 Other natural fibers 48 Fertilizers 88 Other special machinery
9 Crude wood 49 Synthetic resins, fibers 89 Serv industry machinery
10 Fishery 50 Paints,varnishes,lacquers 90 Pumps
11 Iron ore 51 Drugs and medicines 91 Mech handling equip
12 Coal 52 Soap,other preparations 92 Oth non-electrical mach
13 Non-ferrous metal ore 53 Chemical products n.e.c. 93 Radio,TV,phonograph
14 Crude petroleum 54 Petroleum refineries 94 Other telecomm equip
15 Natural gas 55 Fuel oils 95 Household appliances
16 Non-metallic ore 56 Product of petroleum 96 Computers
17 Electrical energy 57 Product of coal 97 Other office machinery
18 Meat 58 Tyre and tube 98 Semiconductors
19 Dairy and eggs 59 Rubber products,n.e.c. 99 Electric motors
20 Presrvd fruits & veget 60 Plastic products,n.e.c. 100 Batteries
21 Preserved seafood 61 Glass 101 Elect bulbs,lighting eq.
22 Veget & animal oils, fats 62 Cement 102 Electrical indl appliance
23 Grain mill products 63 Ceramics 103 Shipbuilding,repairing
24 Bakery products 64 Non-metal prods n.e.c. 104 Warships
25 Sugar 65 Basic iron and steel 105 Railroad equipment
26 Cocoa, chocolate,etc 66 Copper 106 Motor vehicles
27 Food products n.e.c. 67 Aluminum 107 Motorcycles,bicycles
28 Prepared animal feeds 68 Nickel 108 Motor vehicles parts
29 Alcoholic beverage 69 Lead and zinc 109 Aircraft
30 Non-alcoholic beverage 70 Other Non-ferrous metal 110 Other transport eq
31 Tobacco products 71 Metal furn & fixtures 111 Prof measure instruments
32 Yarns and threads 72 Structural metal products 112 Photo,optical goods
33 Cotton fabric 73 Metal containers 113 Watches and clocks
34 Other textile products 74 Wire products 114 Jewelry
35 Floor coverings 75 Hardware 115 Musical instruments
36 Wearing apparel 76 Boilers and turbines 116 Sporting goods
37 Leather and hides 77 Aircraft engines 117 Ordnance
38 Leather products 78 Int combustion engines 118 Works of art
39 Footwear 79 Other power machinery 119 Manufactures n.e.c.
40 Plywood and veneer 80 Agricultural machinery 120 Scraps,used,unclassified
Inforum Spring 2007
The LIFT Model of the U.S. Economy
The Inforum Lift (Long-term Interindustry Forecasting Tool) model is unique among large-scale models
of the U.S. economy. Combining an interindustry (input-output) formulation with extensive use of
regression analysis, it employs a "bottom-up" approach to macroeconomic modeling. For example,
aggregate investment, total exports, and employment are not determined directly, but are computed by
the sum of their parts: investment by industry, exports by commodity, and employment by industry.
Indeed, Lift contains full demand and supply accounting for 97 productive sectors. (See below for Lift
sector titles.)
In short, the demand/production block of Lift uses econometric equations to predict the behavior of real
final demand (consumption, investment, imports, exports, government) at a detailed level. Then, the
detailed predictions for demand are used in input-output production identity to generate gross output
(total revenue adjusted for inflation). Lift's approach to projecting industry prices is similar.
Behavioral equations estimate each value-added component (e.g., compensation, profits, interest, rent,
indirect taxes) for each industry. Value added per unit of output is then combined with the prices of
intermediate goods and services with the input-output price identity to form an indicator for industry
prices. Prices by industry are also dependent on measures of slack in each industry, and, in some cases,
international prices. Thus, income and prices are directly related and are consistent. In turn, relative
price terms and income flows are included as independent variables in the regression equations for final
demand, creating a simultaneity between final demand and value-added.
This bottom-up technique possesses several desirable properties for analyzing the economy. First, the
model works like the actual economy, building the macroeconomic totals from details of industry activity,
rather than distributing predetermined macroeconomic quantities among industries. Second, the model
describes how changes in one industry, such as increasing productivity or changing international trade
patterns, affect related sectors and the aggregate quantities. Third, parameters in the behavioral equations
differ among products, reflecting differences in consumer preferences, price elasticities in foreign trade,
and industrial structure. Fourth, the detailed level of disaggregation permits the modeling of prices by
industry, allowing one to explore the causes and effects of relative price changes.
Despite its industry basis, Lift is a full macroeconomic model, with more than 800 macroeconomic
variables determined consistently with the underlying industry detail. This macroeconomic
"superstructure" contains key functions for household savings behavior, interest rates, exchange rates,
unemployment, taxes, government spending, and current account balances. Like in an aggregate
macroeconomic model, this structure insures that Lift exhibits "Keynesian" demand driven behavior
over the short-run, but neoclassical growth characteristics over the longer term. For example, while
monetary and fiscal policies and changes in exchange rates can affect the level of output in the short-to-
intermediate term, in the long term, supply forces -- available labor, capital and technology -- will
determine the level of output.
Another important feature of the Lift model is the importance given to the dynamic determination of
endogenous variables. For example, investment depends on a distributed lag in the output growth of
investing industries and imports and exports depend on a distributed lag of foreign price changes.
Therefore, Lift model solutions are not static, but are fully capable of projecting a time path for the
endogenous quantities.
Finally, the Lift model is linked to other, similar models with the Inforum Bilateral Trade Model (BTM).
Countries included in this system include the U.S., Japan, China, and the major European economies.
Through this system, sectoral exports and imports of the U.S. economy respond to sectoral level demand
Inforum Spring 2007
and price variables projected by models of U.S. trading partners. In summary, the Lift model is
particularly suited for examining and assessing the macroeconomic and industry impacts of the
changing composition of consumption, production, foreign trade, and employment as the economy
grows through time.
The current model is the fourth discrete version of a modeling framework that has been in continuing
existence since 1967. Since its inception, Lift has continued to develop and change. We have learned
more about the properties of the model through working with clients, and in doing our own simulation
tests. We have learned about the behavior of the general Inforum type of model, from work with our
partners in other countries. Finally, through many experiments, we have learned that many principles of
economics, while attractive theoretically, are difficult to implement practically. We will continue to
experiment, and share ideas, and bring the models closer to our vision of what they should be. A
detailed description of the Lift model can be found at:
http://www.inforum.umd.edu/WorkPaper/INFORUM/wp01002.pdf
Inforum Spring 2007
Producing Sectors of the Lift Model of the U.S. Economy
1 Agriculture, forestry, & fish 45 Elect. lighting & wiring eq 84 Physicians
46 TV's, VCR's, radios 85 Other medical serv & dentists
Mining 47 Communication equipment 86 Nursing homes
2 Metal mining 48 Electronic components 87 Education, social serv, NPO
3 Coal mining
4 Natural gas extraction Transportation Equipment Miscellaneous
5 Crude petroleum 49 Motor vehicles 88 Government enterprises
6 Non-metallic mining 50 Motor vehicle parts 89 Non-competitive imports
51 Aerospace 90 Miscellaneous tiny flows
Construction 52 Ships & boats 91 Scrap & used goods
7 New construction 53 Other transportation equip 92 Rest of the world industry
8 M & R construction 93 Government industry
Instruments & Miscellaneous 94 Domestic servants
Non-Durables Manufacturing 95 Inforum statistic discrepancy
9 Meat products 54 Search & navigation equip 96 NIPA statistical discrepancy
10 Dairy products 55 Medical instr & supplies 97 Chain weighting residual
11 Canned & frozen foods 56 Opthalmic goods
12 Bakery & grain mill product 57 Other instruments
13 Alcoholic beverages 58 Miscellaneous manufacturing
14 Other food products
15 Tobacco products Transportation
16 Textiles and knitting 59 Railroads
17 Apparel 60 Truck, highway pass transit
18 Paper 61 Water transport
19 Printing & publishing 62 Air transport
20 Agric fertilizers & chemicals 63 Pipeline
21 Plastics & synthetics 64 Transportation services
22 Drugs
23 Other chemicals
24 Petroleum refining Utilities
25 Fuel oil 65 Communications services
26 Rubber products 66 Electric utilities
27 Plastic products 67 Gas utilities
28 Shoes & leather 68 Water and sanitary services
Durable Material & Products Trade
29 Lumber 69 Wholesale trade
30 Furniture 70 Retail trade
31 Stone, clay & glass 71 Restaurants and bars
32 Primary ferrous metals
33 Primary nonferrous metals Finance & Real Estate
34 Metal products 72 Finance & insurance
Non-Electrical Machinery 73 Real estate and royalties
35 Engines and turbines 74 Owner-occupied housing
36 Agr., constr., min & oil equip
37 Metalworking machinery Services
38 Special industry machinery 75 Hotels
39 General & misc. industrial 76 Personal & repair services
40 Computers 77 Professional services
41 Office equipment 78 Computer & data processing
42 Service industry machinery 79 Advertising
80 Other business services
Electrical Machinery 81 Automobile services
43 Elect. industry equipment 82 Movies & amusements
44 Household appliances 83 Private hospitals
Inforum Spring 2007