1
Evaluating the Regional Economic Effects of
Structural Funds Programs Using the
REMI Policy Insight Model
By1
Frederick Treyz, Ph.D
and
George Treyz, Ph.D
1
The authors are respectively CEO and President of Regional Economic Models, Inc.; 306 Lincoln Ave.
Amherst, MA 01002; (tel) 413-549-1169; info@remi.com.
The results and analysis of Southern Italy represent the joint work of REMI and IRPET. The staff who are
actively involved in the analysis of IRPET are Sara Mele, Roberto Pagni and Renato Paniccia.
2
Evaluating the Regional Economic Effects of Structural Funds Programs Using the
REMI Policy Insight Model
By Frederick Treyz, Ph.D., George Treyz, Ph.D.2
This paper presents scenarios to address major types of Structural Funds programs
using the REMI macroeconomic model. The purpose is to show how the total socioeconomic effects of a broad range of programs can be evaluated using a widely available
model that is developed to analyze the economies of nations and sub-national regions.
The model results show the total macroeconomic impacts of changes that occur due to
direct policy effects.3
We show the total effect of structural measures and public investments on a
representative economy. The analysis shows how to conduct a comparative evaluation of
the future effectiveness of expenditures for different investments and structural and
cohesion funds programs. These comparisons will provide a basis for developing
information necessary for policy-makers to predict the relative effectiveness of various
expenditures on major economic development goals. This facilitates the process of
making choices about the future allocation of operation program funds.
Here we provide specific analyses and results for representative structural-funds
analysis covering a broad range of programs. These programs include subsidies for the
2
The results and analysis of Southern Italy represent the joint work of REMI and IRPET. The staff who
are actively involved in the analysis of IRPET are Sara Mele, Roberto Pagni and Renato Paniccia.
3
This model has been in continual development since 1980 and is well-documented in academic literature.
See for example, Fan, Treyz and Treyz (2000), Greenwood, Hunt, Rickman and Treyz (1991), and Treyz,
Rickman and Shao (1992).
3
productive sector, investments in transportation infrastructure, human-capital investment
and tourism investment.
The approach is to represent the direct effects of a policy, use these effects as
inputs into a macroeconomic model, and generate the economic and demographic
changes that occur as a result of the policy. The simulation results are reported as the
difference in economic activity on an annual basis that is caused by the direct economic
change.
For example, investing in transportation infrastructure increases the quality of
transportation infrastructure, which has the effect of reducing transportation costs. The
direct effects of the transportation investment are based in part on an estimate of the
reduction in percentage terms of transportation costs. These transportation cost changes
are entered into the Transportation Cost Matrix of the Policy Insight model as a change in
travel between the regions in a multi-regional model. From this, the results of the
analysis show the macroeconomic effects of transportation improvements.
Particular attention is focused on supply-side effects such as those of subsidies
for:

Major manufacturing industries

Investment in labor productivity

Investment due to an increase in access to intermediate inputs

Reduction in transportation costs due to an investment in transportation
infrastructure
The results are stated in terms of changes in key macroeconomic and
demographic indicators. The inputs for the subsidies to industries, for example, have
4
major effects on exports, intermediate and final sales, exports to the rest of the nation,
output, gross domestic product, employment, unemployment rate, migration, relative
production costs, and real disposable income per capita.
For example, consider the analysis of a subsidy. The subsidy may lead to a longterm location of a firm due to the relative sensitivity in the locations of fixed capital stock
to the initial investment cost. Shares of the local and export markets would increase
output. Employment would be increased and unemployment decreased as output
increases. The employment and real-wage gains would decrease the outward net
migration that was indicated in the baseline forecasts. The model continues to track
through all of the causality until a simultaneous solution for each year is found.
In order to examine structural change, it is necessary to have a model that
explains all of the key cause-and-effect relationships in the economy so that interventions
can be introduced to change the status quo. For example, if a model does not include
labor supply by occupation and/or industry, it is difficult to predict the effects among
training programs. Models that do not explicitly include the effects of improved access
to labor or intermediate inputs will miss one of the primary reasons that some regions
prosper, which is the availability of a large pool of trained labor and access to specialized
intermediate inputs. In a draft paper that we are preparing for the European Commission,
we conclude that “each of the models represents a different focus. Quest II focuses on
forward-looking rationality and international trade balance, Venables and Gasiorek on
micro-economic behavior of firms in space, REMI on explicit structures based on
maximizing behavior of actors in response to current conditions and key statistical
parameters based on large data sets, and HERMIN on special characteristics in different
5
countries as revealed in the relationships in recent years between variables in aggregate
time series estimates.”4
The major goal for E.U. Structural Funds and many other Public Investments is
sustainable economic development. Unfortunately, accurate estimates of the relative
effectiveness of alternative public expenditures in meeting this goal are often not
available. Thus, allocation of euros may or may not be optimal.
In order to determine the relative effectiveness three steps are required.
1. Examine each of the operational plans in detail. In particular, we must
examine the direct effects on the economy in the short and long run. While
the direct-demand effects will be important in the short run, it is the long-run
supply-and-demand effects that will be the most important factors in
sustainable development.
2. Input these direct effects into a structural, economic-policy-analysis model
that includes all of the key chains of causality through which public fund
investments will influence the economy.
3. Run the model and calculate the relative effectiveness of the euro expenditure
in accomplishing the major objectives of the public investments.
Some examples of the ways that public policy can directly affect the socioeconomic outcomes in the economy are:
Increase the private capital investments in the economy:
Instruments:
 Subsidies
 Tax policies
 Facilitate the working of capital markets
“A Comparison of the Four Economic Policy Analysis Models that Have Been Selected by the E.U.
Commission for Structural Fund Evaluation.” By Frederick Treyz and George Treyz. REMI mimeo.
4
6
Increase demand on a permanent basis:
Instruments:
 Public investments in tourism
 Infrastructure investments
Reduce Labor and Capital Costs:
Instruments:
 Reduce non-wage labor costs
 Tax policies
Increase Productivity:
Investments:
 Technology development
 Labor training
 Increase access to labor and intermediate inputs
 Transportation infrastructure
Make the area a more attractive place to live in:
Investments:
 Public infrastructure
 Environmental improvements
 Reduce crime
 Improve consumer access to goods and services
 Tax policy
Increase Labor Supply:
Investments:
 Effect the age structure of the population
 Increase participation rate
 Change tax support for non-workers
 Change attitudes; enable people over 65 to continue to work
 Promote local products
 Add variety by expanding some industries
7
The model used must be able to forecast the effects on the total regional economy
of the direct inputs into the model represented by structural funds intervention. It must
also have a plausible structure that captures all of the key chains of causality through
which the public investment (or micro-analysis) directly affects flow through the
community. We believe that the REMI model is the only model that meets these
requirements. The remainder of this paper will be devoted to a description of and
preliminary results from two models that we have built for the E.U. Directorate-General
for Regional Policy Coordination of Evaluation Commission. One model was jointly
developed and applied by REMI and IRPET (Instituto Regionale Par La Programmoz
oine Economica Toscana, Via La Farina 27, 50132 Florence, Italy). The other REMI
Policy Insight model is for a multi-regional configuration of four areas in Spain.
In order to build our structural model of Italian regions we gathered data for all
areas and then built a database that included raw data such as wage rates, employment,
and detailed demographic data. In addition, using our model structure we were able to
generate productivity and other concepts.
8
Figures 1-5 are maps of Italy that represent economic effects on the textile
industries. Such maps can be helpful in assessing the strengths and weaknesses of a
region. Figure 1 shows the industry of textiles and textile products. It shows that
comparative wage rates are highest in Piemonte in Northern Italy.
Italy: Wage Rate: Textiles & Textile Products
Figure 1
9
Figure 2 illustrates labor productivity as we estimate it using the comparative
availability of workers to choose from in the textile industry and in occupations used in
that industry. In this case, Lombardia has the highest productivity.
Italy: Labor Productivity: Textiles & Textile
Products
Figure 2
10
Figure 3 shows wages adjusted for productivity.
Italy: Wage Rate divided by Labor
Productivity: Textiles & Textile Products
Figure 3
11
Figure 4 shows the price of textiles considering both the production cost and the
access to variety.
Italy: Composite Price: Textiles & Textile
Products
Figure 4
12
Figure 5 shows us that production costs are lowest in Lombardia despite their
relatively high wage rates.
Italy: Production Cost: Textiles & Textile
Products
Figure 5
This background can illustrate the structural detail in the REMI model and could
be used along with similar data for other industries to select policies to develop as
economic development strategies. An overview of the REMI model abstracted from the
REMI model documentation book is included as an appendix to this paper. Further
information about the model and its parameter estimation, etc., is a forthcoming OECD
publication.5
Using a REMI-IRPET model of Objective 1 areas in Southern Italy, IRPET and
REMI have carried out a number of simulations for selected structural fund expenditures.
13
Here we will first present the preliminary comparative results of these simulations and
then we will describe the inputs that went into REMI Policy Insight, as well as the
explanations of the macro results for each step in the simulation process.
In order to evaluate the effectiveness of the alternative structural-funds program,
the policy simulation model has made a comparison of benefit per 100,000 euros of
alternative benefit measures. Table 1 shows four different types of investment.
2006
Jobs per
100,000
euros
Investment in 50%
Subsidy to AG
Equipment
Human Capital: Blue
Collar Training
Tourism Productivity
Improvements and
Promotions
Integrated Transport
System
2006
2006
Labor force Population
per 100,000 per 100,000
euros
euros
2006
GDP per
euro
invested
2006
Real
Disposable
Income per
euro
32
8
2
1
0.3
5
1
0
0.3
0.1
24
8
2
1.5
1.2
19
5
1
0.6
0.2
Table 1: Preliminary Estimates of the Effects of Structural Fund Investments in the Objective 1 areas of
Southern Italy in 2006.
The first simulation was for a program that subsidized agricultural equipment
purchases. The first three results of the simulation are stated per 100,000 of the
expenditures during the seven-year period from 2000-2006. The first result shows that 32
jobs were generated in 2006 from the past and current subsidies for the equipment. In
2006, some of the results are due to the supply-side productivity gains from investments
“The Evaluation of Programs Aimed at Local and Regional Development: Methodology and Twenty
Years of Experience Using REMI Policy Insight.” By Frederick Treyz, and George Treyz. Forthcoming
OECD publication.
5
14
in previous years as well as the demand-side effects of spending for the seventh year of
the equipment in 2006.
The first is subsidy to agricultural equipment. The second policy is subsidy for
human capital, in particular blue-collar training. The third is a study on tourism, and the
fourth is transportation analysis. In each of these cases, benefit measures were used,
including jobs, labor force, and population per 100,000 euros. Gross domestic product
(GDP) per euro and real disposable income per euro generated by the various structural
funds programs were also used in the models.
Looking from one policy to the others we see that in terms of job creation there is
a big difference between various policies. For example, investments in blue-collar
training have developed relatively few jobs per euro, whereas a subsidy in agricultural
equipment generates a high number of jobs. There are similar differences, although not
quite as wide-ranging, in terms of the other measures of changes in population, GDP, and
disposable personal income.
The subsidy to agricultural equipment generated the highest number of jobs per
euro since this subsidy provides an indirect subsidy to the agricultural industry.
Agriculture is a relatively labor-intensive industry. Thus by increasing the
competitiveness of agriculture through the subsidy there are many jobs created. Even in
the case of GDP per euro this effect is relatively high. However, since agriculture is a
relatively low-income industry, the effects on real disposable income creation per euro of
subsidy are not much higher than they are for the blue-collar training or transport system
investment. In fact, they are significantly lower than they are for the improvements in
15
tourism productivity. This simulation did not consider possible additional costs to the
E.U. related to additional agricultural surpluses.
To measure the jobs generated by the subsidy to agricultural equipment in 2006,
we first measured the amount of agricultural equipment that would be new equipment.
Then we assumed that one-third of the new equipment would have been purchased
without the subsidy. Next we divided 60% of the equipment for agriculture by the
capital-to-output ratio in order to give us an estimate of the amount of extra output we
would have for agriculture. Then 80% of that extra output was put into sales that would
occur without competition to other firms in the area. This is particularly true because of
the assumption that the government of the E.U. absorbs the surplus production. We
assumed that 20% of the surplus firm output would be in competition with other firms in
the area. We also assumed that the life expectancy of the equipment would be 10 years.
These gave us the inputs for agricultural product sales. The other 40% of the extra
equipment was attributed to the food and beverage industry and was divided by the
capital ratio of .95, giving us the amount that was spent in the food-products industry.
All of the food sold in the food-product industry is in competition with other food sales
and was thus entered into the model as firm sales. This accounts for any other firm in the
food-products industry in Southern Italy, as well as extra sales due to a greater variety of
suppliers. We calculated the share of local contribution to the subsidy and we added that
to personal taxes. The last input was the amount of expenditure per year on the
machinery equipment that was purchased each year. This was put in as demand
originating within the area but supplied within and without the area. The extra demand
supplied locally added to economic activity through 2006 when the program ended. This
16
completed the inputs that were put into the final total for the expenditures in agricultural
equipment.
The effects of the subsidy consist of four parts. The final demand for machinery
and agricultural equipment in 2006 was 7.4% of the final total employment change. The
sales of agricultural goods consisted of 70.4%. The effect of food and beverage product
sales made up 22.7% of the final total. Yet, 0.5% of the total employees are taken away
due to extra local taxes to pay the local portion of the subsidy.
2012
Jobs per
100,000
euros
Investment in 50%
Subsidy to AG
Equipment
Human Capital: Blue
Collar Training
Tourism Productivity
Improvements and
Promotions
Integrated Transport
System
2012
2012
Labor force Population
per 100,000 per 100,000
euros
euros
2012
GDP per
euro
invested
2012
Real
Disposable
Income per
euro
14
8
4
0.5
0.2
6
1
0
0.5
0.2
25
13
6
1.7
1.4
8
5
2
0.3
0.1
Table 2: Preliminary Estimates of Structural Fund Investments in the Objective 1 areas of Southern Italy in
2012.
In 2012, the jobs generated by the subsidy to agricultural equipment decrease.
These results are supply-side effects from the equipment. This is due to the fact that the
agricultural equipment has a 10-year life expectancy. Therefore, the new machinery
wears out, causing a loss of employees. Jobs per 100,000 euros decline due to the aging
of the equipment.
17
The final demand for machinery and agricultural equipment in 2012 was –3.4% of
the final total. The sales of agricultural goods made up 75.3% of total. The effect of
food and beverage sales consisted of 24.3%. The total employees due to extra local taxes
consist of 0.4% of the final total.
SIMULATION: Occupational training in Regione Campania, data from P.O.R.
(Operative Regional Program) Ob. 3 2000-2006 (Sara Mele).
Policy variables: assumptions and sources
Occupational training, detail: Blue collars/Other personnel
We have estimated the number of trainees (blue collars/other personnel)
considering an average cost per person of about 6.000 euros. This training increased
labor access and thus labor productivity. This increase in labor productivity in turn led to
lower cost in Southern Italy, which increased competitiveness and thus sales. It also
reduced local cost for locally produced goods. In calculating labor access, we assumed
that only 50% of the trained labor force would remain in the area and be active in the
labor force. This assumption was made based on the high unemployment rates in
Southern Italy. This effect contributed 15% of the employment effect in 2006 and 111%
in 2012.
Exogenous Final Demand, detail: Education
We assumed that training programs are offered by private agencies in the
education sector. Therefore, we introduced an exogenous demand for these firms equal
18
to the total cost of training courses (completely financed by public resources). This effect
contributed 92% to the employment effect in 2006 and –13% in 2012. The negative
effect in 2012 (six years after the training would be completed) was due to the
stimulation to the economy in the first six years, which led to residential and nonresidential construction, which was in greater supply in 2006 than it otherwise would
have been.
Personal Taxes, detail: Applicable Personal Income
Priority III for “Human resources” is financed by the European Social Fund
(63%) and national (26%) and regional (11%) governments. So we have estimated the
increase in personal taxes for citizens of Regione Campania necessary to finance the
Regional Operational Programme.
Regional co-financing is completely collected with regional taxes. For the
national co-financing we have calculated the share of national taxes paid by Campania’s
citizens. Personal taxes from 2000-2006 reduced the result in 2006 and increased it by
9% in 2012 using the same effect as explained for the training expenditure above.
SIMULATION: Subsidies to increasing tourism sector competitiveness and tourism
expenditure (Operative Regional Program) Ob. 3 2000-2006.
Total Factor Productivity
The main objective of the measure is to increase Total Factor Productivity (TFP).
In particular the target is expressed in terms of Value added per unit of labor in the
19
tourism sector (Hotels and Restaurants). The present value is 48.0, while the target value
is 55.0. The TFP of the Hotel and Restaurant industry has been increased consistently.
The increase in productivity had an important effect on the cost of production in the
Hotel and Restaurant industry. For local residents this had the effect of reducing costs of
restaurant meals and thus increasing real incomes and also leading to some substitution of
restaurant meals for all items that they might otherwise have purchased. The high local
content of restaurant meals added to the employment gain. Of course, some reduction of
capital and labor input per meal mitigated this effect. The other major effect was the
reduction in the cost of Hotels and Restaurants in Southern Italy for both Italian tourists
and tourists from abroad. Both of these effects increased the “export” of these services
and thus increased employment. In this particular simulation we used a depreciation rate
of 5% for the depreciation of the equipment.
Exogenous demand in Investments
The increasing investment in Hotels and Restaurants by producing sectors have
been estimated through the investment bridge matrix associated to Campania I-O matrix
estimated by IRPET.
SIMULATION: Subsidies to inter-modal commodities transportation (Operative
Regional Program) Ob. 3 2000-2006.
20
Policy variables: assumptions and sources
Transportation Cost
One of the main indicators of such measures is to increase the amount of tons of
commodities by railroad per 100 inhabitants. The present value is 1.99 and the target
value is 2.75. This will determine a decrease in transportation cost by moving
commodities from road to railroad (the parameter has been estimated using data drawn
from the Transportation National Account 1994 and Italian State Railways Yearbook).
The projection of decreasing transportation cost at the regional level has been estimated
using the transport margins in the Italian Objective 1 I-O matrix at market price. This
gain contributed 29.6% of the gain in employment in 2006 and 76.2% in 2012.
Accessibility
Increase in accessibility adds to the variety of intermediate production inputs that
producers can choose from. The accessibility costs in 2006 are 14% of the employment
gains, and in 2012 this gain is raised to 65.9%. These intermediate inputs range from
access of components of products that are produced, to the access of specialized legal
services. This greater access creates an agglomeration effect that increases productivity
and reduces costs. These reductions improve competitiveness and increase sales and
employment.
Exogenous demand in Construction
The increasing demand in investment in Construction has been estimated through
the total amount of investment and the usage of the specific intermediate column cost of
21
railroad building provided by the Ministry of the Economy. The construction of the
infrastructure is 55.9% of the employment in 2006, but reduces the gain by -42.2% in
2012. Again the reduction in 2012 is an echo of the extra construction of residential and
commercial space simulated by the accelerated economic activity from 2000-2006.
An example of a 1% reduction in transportation costs for Andalucia
This simulation for Andalucia is based on the arbitrary assumption that a project
is undertaken to reduce transportation costs by 1%. Figure 6 shows a reduction of
transportation cost of internal transportation in Andalucia to 99% from what they are in
the baseline. The model used is a four-area model that would allow us to calculate input
and reductions of transportation costs to any of the other areas in the model.
Policy Variable Values for
Transportation Costs down 1% for
Andalucia
Figure 6
22
Figure 7 assumes an access gain of one-half the value of transportation costs since
access depends on transportation costs and other factors.
Policy Variable Values for
Accessibility Costs down .5% for
Andalucia
Figure 7
Figure 8 indicates a 0.05% reduction in labor costs. In this case, this represents an
average reduction of commuter time by 15 seconds each way. If commuting time is
valued at one-half of the wage rate this would be a 0.05% reduction in the premium the
average commuter would add to his or her reservation wage.
23
Policy Variable Values for
Commuting Costs down .05% in
Andalucia
Figure 8
Figure 9 shows reductions in the cost of production for all industries except the
Finance, Insurance and Real Estate industry. The reason that this industry is an exception
is that real estate and housing prices will be up due to increased demand resulting from
incomes and populations that are higher than they were in the baseline.
Transportation Costs –1%, Accessibility Costs -.5%,
and Commuting Costs -.05% for Andalucia
Compared to REMI Standard Regional Control
Rel Cost of Production by % Change for Andalucia
Figure 9
24
Figure 10 illustrates the increase in employment that occurs due to lower prices
and increased market shares that occur because of increases in competitiveness.
Transportation Costs –1%, Accessibility Costs -.5%,
and Commuting Costs -.05% for Andalucia
Compared to REMI Standard Regional Control
Total Employment by Percent Change for Andalucia
Figure 10
Figure 11 shows the output of these same effects. The self-supply effects are due
to the immediate reduction in costs and prices in the home market while the export
response to more competitiveness grows more gradually.
Transportation Costs –1%, Accessibility Costs -.5%,
and Commuting Costs -.05% for Andalucia
Compared to REMI Standard Regional Control
Output Components by Percent Change for Andalucia
Figure 11
25
Next we show the changes in market shares in 2010 for a sample industry in
Figure 12. Here we note that the Andalucians’ share of its own market increases by
0.23%. The Andalucians’ share of the markets in other areas of Spain increase by
between 0.03% and 0.04%, while its increase in sales to the rest of the world goes up by
0.07%. These changes lead to a total increase of 0.18% in output for Andalucia. Note
also that its purchases from other Spanish areas go down by 1.8% as the increased
competitive effect offsets the stimulated effect of increase in the Andalucian economy.
Transportation –1%, Accessibility Costs -.5%,
and Commuting Costs -.05% for Andalucia
Compared to REMI Standard Regional Control
Domestic Trade Flows by Percent Change
Figure 12
Table 3 shows the chain of causality even more completely. Note that the GDP
gain remains even though the employment gains are lower in 2017 than in 2000. This is
due to the general gain in the productivity of labor assumed in the baseline as well as
increasing labor productivity due to increased access. The improved conditions in
Andalucia reduce the number of the out-migration in the baseline forecast. At first,
exports to other areas decrease and it is only over time that the gains in competitiveness
increase the penetration into those markets.
26
16-Jun-03
REMI Standard Reg Control
Accs -.5% &trans cost down by 1%& icom cst dn .05%
[Top] - Differences
Andalucia
Variable
2000
Employment (Thous)
5.701
Gross Regional Product (GRP) (1,000M 97Euro)
0.2426
Personal Income (1,000M Nom Euro)
0.1718
PCE-Price Index (97 Euro)(Reg vs National Baseline) -0.01671
Real Disp Personal Income (1,000M 97Euro)
0.1946
Population (Thous)
0.3516
Econ Migrants
0.1204
Total Migrants
0.1204
Labor Force
1.509
Demand (1,000M 97Euro)
0.3862
Output (1,000M 97Euro)
0.4134
Delivered Price
-0.0002
Relative Cost of Production
-0.00011
Labor Intensity
-1.55E-05
Labor Access Index
0.000342
Indust Mix Index
0
Reg Pur Coeff (SS over Dem)
0.000784
Imports (1,000M 97Euro)
-0.0282
Self Supply (1,000M 97Euro)
0.4143
Exports to Multiregions (1,000M 97Euro)
-0.00345
Exports to Rest of Nation (1,000M 97Euro)
0
Export to Rest of World (1,000M 97Euro)
0.002526
Wage Rate (Thous NomEuro)
0.007013
2010
3.927
0.2654
0.2077
-0.05173
0.2071
1.657
0.1106
0.1106
2.277
0.3037
0.392
-0.00039
-0.00031
-8.60E-05
0.000619
0
0.000811
-0.06997
0.3736
0.005084
0
0.01323
0.01286
2017
3.543
0.2836
0.2607
-0.07393
0.2251
2.539
0.1014
0.1014
2.43
0.2983
0.4135
-0.00044
-0.00036
-0.00012
0.000627
0
0.000846
-0.08693
0.3852
0.00839
0
0.01989
0.01509
Table 3: The Effects of 1% Reduction in Transportation Costs in Andalucia, including an Access Increase
of .5% and a Commuting Cost Decrease of .05%
This is a simplified example. In a real example we would have to consider the
building of the transportation infrastructure, the cost of maintenance and the source of the
funding of the project. Also, if this project increases the transportation costs to and from
other regions this would reduce the costs of other areas’ deliveries into Andalucia, in
addition to Andalucia’s reduction of costs into these areas. In some industries this might
increase Andalucian output, and in other areas it might reduce it.
27
Summary
In this paper we have shown how looking at returns to structural-fund investments
would increase the efficiency of the investment of these funds in achieving their
objectives. We have also made the case for using a model that fully represents the
structure of the economy to show the macro-economic effects of investments intended to
change the structure of the local economy.
28
APPENDIX: Overview of the REMI Model6
OVERVIEW OF THE MODEL
REMI Policy Insight is a structural economic-forecasting and policy-analysis model. It
integrates methods from several other model types to help its users understand the total macroeconomic effects of policy decisions. It has the inter-industry detail available from input-output models.
It allows for behavioral responses to housing and consumer prices, wages, and production costs as in
computable general equilibrium models. Dynamic responses and other parameters are estimated using
econometric methods. The agglomeration economies that are key to a region are represented using new
economic geography theory. The model is dynamic, generating forecasts and simulations on an annual
basis, and accounting for behavioral responses to wage, price, and other economic factors.
While the REMI model comprises thousands of simultaneous equations, its structure is
relatively straightforward. The exact number of equations varies depending on the extent of detail in the
model. The overall structure of the model can be summarized in five major blocks: (1) output and
demand, (2) labor and capital demand, (3) population and labor force, (4) wages, prices and costs, and
(5) market shares. The blocks and their key interactions are shown in Figures 1 and 2.
Figure 1
6
“The REMI Economic Geography Forecasting and Policy Analysis Model.” Treyz and Treyz. (2002).
29
Figure 2
The output block consists of output, demand, consumption, investment, government
spending, exports, and imports, as well as feedback from output change due to the change in the
productivity of intermediate inputs. The labor and capital demand block includes labor intensity and
productivity as well as demand for labor and capital. The demographic block includes labor force
participation rate and migration equations. The wages, prices and production costs block includes
composite prices, determinants of production costs, the consumption price deflator, housing prices, and
the wage equations. The market shares block includes the proportion of local, inter-regional and export
markets captured by each region.
Models can be built as single-region, multi-region, or multi-region national models. A region is
defined broadly as a sub-national area, and could consist of a state, province, county or city, or any
combination of sub-national areas. Within a large, multinational currency zone such as the European
Union, models of a national economy can be built using the same economic framework employed in
regional models.
Single-region models consist of an individual region, called the home region. The rest of the
nation is also represented in the model. However, since the home region is only a small part of the total
nation, the changes in the region do not have an endogenous effect on the variables in the rest of the
nation.
30
Multi-regional models have interactions among regions, such as trade and commuting flows.
These interactions include trade flows from each region to each of the other regions. There are also
multi-regional price and wage cost linkages. The average price in each of the regions depends on the
production cost in each of the other regions plus the transportation to the region in question.
Block 1. Output and Demand
This block includes output, demand, consumption, investment, government spending, import,
product access and export concepts. Output for each industry in the home region is determined by
industry demand in all regions in the nation, the home region’s share of each market, and international
exports from the region.
For each industry, demand is determined by the amount of output, consumption, investment
and capital demand on that industry. Consumption depends on real disposable income per capita,
relative prices, differential income elasticities and population. Input productivity depends on access to
inputs because with a larger choice set of inputs, it is more likely that the input with the specific
characteristics required for the job will be formed. In the capital stock adjustment process, investment
occurs to fill the difference between optimal and actual capital stock for residential, non-residential, and
equipment investment. Government spending changes are determined by changes in the population.
Block 2. Labor and Capital Demand
The labor and capital demand block includes the determination of labor productivity, labor
intensity and the optimal capital stocks. Industry-specific labor productivity depends on the availability
of workers with differentiated skills for the occupations used in each industry. The occupational labor
supply and commuting costs determine firms’ access to a specialized labor force.
Labor intensity is determined by the cost of labor relative to the other factor inputs, capital and
fuel. Demand for capital is driven by the optimal capital stock equation for both non-residential capital
and equipment. Optimal capital stock for each industry depends on the relative cost of labor and capital,
and the employment weighted by capital use for each industry. Employment in private industries is
determined by the value added and employment per unit of value added in each industry.
Block 3. Population and Labor Force
The population and labor force block includes detailed demographic information about the
region. Population data is given for age, gender and ethnic category, with birth and survival rates for
each group. The size and labor force participation rate of each group determines the labor supply. These
participation rates respond to changes in employment relative to the potential labor force and to
changes in the real after-tax wage rate. Migration includes retirement, military, international and
economic migration. Economic migration is determined by the relative real after-tax wage rate, relative
employment opportunity and consumer access to variety.
31
Block 4. Wages, Prices and Costs
This block includes delivered prices, consumer prices, housing prices, production costs,
equipment costs, the consumption deflator, and the wage equation. Economic geography concepts
account for the productivity and price effects of access to specialized labor, goods and services.
These prices measure the price of the industry output, taking into account the access to
production locations. This access is important due to the specialization of production that takes place
within each industry, and because transportation and transaction costs of distance are significant.
Composite prices for each industry are then calculated based on the production costs of supplying
regions, the effective distance to these regions, and the index of access to the variety of output in the
industry relative to the access by other uses of the product.
The cost of production for each industry is determined by labor costs, capital costs, fuel costs,
and intermediate inputs. Labor costs reflect a productivity adjustment to account for access to
specialized labor, as well as underlying wage rates. Capital costs include costs of non-residential
structures and equipment, while fuel costs incorporate electricity, natural gas and residual fuels.
The consumption deflator converts industry prices to consumption commodity prices. For
potential migrants, the consumer price is additionally calculated to include housing prices. Housing price
changes depend on changes in income and population density.
Wage changes are due to changes in labor supply and demand conditions, and changes in the
national wage rate. Changes in occupational demand and employment opportunities relative to the labor
force determine wage rates by industry.
Block 5. Market Shares
The market-shares equations measure the proportion of local and export markets that are
captured by each industry. These depend on relative production costs, the estimated price elasticity of
demand, and effective distance between the home region and each of the other regions. The change in
share of a specific area in any region depends on changes in its delivered price and the quantity it
produces compared with the same factors for competitors in that market. The share of local and
external markets then drives the exports from and imports to the home economy.
32
REFERENCES
An Evolutionary New Economic Geography Model; Wei Fan, Frederick Treyz and
George Treyz; Journal of Regional Science, Vol. 40(4), 671-695; 2000.
Migration, Regional Equilibrium, and the Estimation of Compensating Differentials;
Michael Greenwood, Gary Hunt, Dan Rickman, George Treyz; American Economic
Review, Vol. 81(5), 1382-1390; 1991.
The REMI Economic-Demographic Forecasting and Simulation Model; G.I. Treyz, D.S.
Rickman and G. Shao; International Regional Science Review 14(3), 1992:221-253.
The REMI Geography Forecasting and Policy Analysis Model; Frederick Treyz and
George Treyz; Copyright 2002.
“A Comparison of the Four Economic Policy Analysis Models that Have Been Selected
by the E.U. Commission for Structural Fund Evaluation.” By Frederick Treyz and
George Treyz. REMI mimeo.
“The Evaluation of Programs Aimed at Local and Regional Development: Methodology
and Twenty Years of Experience Using REMI Policy Insight.” By Frederick Treyz, and
George Treyz. Forthcoming OECD publication.
“REMI Model Documentation: REMI Policy Insight.” Regional Economic Models,
Inc. Copyright 2003.