Handbook on Urban Economic Base Analysis
Part IV
Analytical Tools and Techniques of Economic Base Analysis1
There are a number of crude research procedures available to assist in explaining the nature of the local economy and its relationships with broader economies. The ability of these procedures to explain the local economy and its relationships is highly dependent upon two things: the quality of the information available, and the quality of the local knowledge that must be used to interpret and verify the results of any quantitative method. The results of these research procedures should thus be used with a great deal of caution.
The most common methodology used for local economic analysis is economic base analysis. Economic base analysis is founded upon the distinction between basic and nonbasic economic activities. Simply explained, this analysis states that the existence and growth of an economy depends upon basic economic activities that produce and sell their goods or services to nonresidents. The more this happens, the more wealth is generated locally and the greater the stability and growth of the local economy. Nonbasic activities produce and sell their goods or services to local residents.
Basic activities are frequently referred to as export activities, urban growth activities, or primary activities. Nonbasic activities are sometimes described as service activities, town fillers, or secondary or auxiliary activities. A simple way to understand the difference between basic and nonbasic activities is to ask the question: "Does this economic activity bring money from outside into the city, or does it re
circulate money within the city?" Activities resulting in a net inflow of money to the city are basic activities; those that do not are nonbasic.
Using this framework, one can see that the term "export" refers to the place a good or service is consumed, rather than whether new money is brought into the city. To simply use the term "exports" as a definition for basic activities would exclude activities that are consumed within the city by nonresidents. For example, island economies can grow and prosper on money brought into their economies by nonresident tourists without having to export anything. Tourism is a basic activity that brings money into an economy, even though the actual consumption of goods or services occurs within (rather than outside) the local economy.
Also inherent in the concept of economic base analysis is an assumption of a cause-and-effect relationship between basic and nonbasic activities. The assumption is that as money comes into the local economy through a basic activity, it is then available for recirculation within the local economy through nonbasic activities. Growth could not occur by a simple recirculation of the same money, but only results from a net inflow of money to the economy. Nonbasic activities are wholly or partly a function of, or dependent upon, the presence of basic activities. Growth or decline in basic activity results in growth or decline in nonbasic activity; this relationship between basic and nonbasic activity is described by a basic to nonbasic ratio.
Research procedures used in economic base analysis therefore attempt to categorize all economic activities as either basic or nonbasic. A major difficulty in economic base analysis is this
classification requirement, as many economic activities result in outputs that are partially consumed by residents and partially consumed by nonresidents. In fact, relatively few economic activities in any city could be considered purely basic or nonbasic.
One of the products of an economic base analysis is a basic to nonbasic ratio. If employment is used as the measurement, a basic to nonbasic ratio of 1:1 for a city would suggest that for each job created in a basic activity, one more job in nonbasic activities would be created, for a total of two jobs.
In general, an economic base analysis can produce the following types of information:
* broad indicators, such as changes in employment by industry;
* basic to nonbasic activity ratio;
* employment mix compared to other cities and/or to the national economy;
* the relative concentration of employment (or coefficients of specialization) by economic activity compared to other cities and/or to the national or international economies;
* the part of employment change that can be attributed to changes in national employment; and
* the part of employment change that can be attributed to changes in local economic activity.
Economic base analysis can provide a deep understanding of the local economy and its position in relation to the national economy and the economies of other cities and countries, especially those of the closest competitors.
Economic Base Analysis: Preliminary Considerations
Most research procedures used in economic base analyses use indirect approaches (statistical data) as opposed to direct approaches (collection of data by a survey designed for the study of the economy in question). Consequently, the results derived using the indirect approach are relatively crude, and should only be used by economic researchers with a thorough knowledge of the local economy and of the applied research procedures.
1Many of the concepts, methodologies, and procedures explained in this part are based on the work of Tiebout (1962); Bryant and Preston (1988); and Bryant, Preston, and Dudycha (1988, 1989, and 1990).
Before conducting an economic base analysis, there are several preliminary decisions required with respect to the parameters and methodologies to be employed:
* selecting a study area,
* determining what will be measured and the units of measurement,
* determining the number of classes and subclasses of economic activity that are appropriate for the study area,
* selecting a methodological approach (direct or indirect), and
* determining the analytical tools to be used for the economic base analysis.
Selecting a Study Area
The selection of the limits of an area to study would at first appear to be a relatively simple matter, but this must be done with care. Slight differences in the area chosen may produce different analysis results. The delineation of the study boundaries must also be done in conjunction with the selection of what will be measured (employment, sales, etc.) and of the analytical tools to be used. The area selected could reflect legal or political boundaries, the developed urban area, the urban area plus its commuting zone, or the retail trade area of the city.
The strongest influence on what area to select will be data availability for the measurements desired. If data are not available for the selected area, either the area will have to be modified to match that for which data are available, or a special data collection will be required (which can be time consuming and expensive). Data availability for urban areas is a problem in the
transitional economies of Central and Eastern Europe, with time-series data simply being unavailable in most cases. While the required data are occasionally available for major cities, they are rarely available for smaller urban areas.
The economic researcher thus must carefully select the area, considering both political and analytical implications. The area chosen will seldom be a perfect match for the needs and could require special manipulations to approximate the desired study area. The selection of the study boundaries could affect the strength or weakness of the economic base analysis.
Determining What Will Be Measured and the Units of Measurement
There is, unfortunately, no one single ideal measurement for use in economic base analysis. Several different measurements can be used, however, each with advantages and drawbacks. The most frequently used measurement is employment, as it usually has the most complete data set available. Other measurements, if they are available, are very useful for verifying the validity of the analysis and adding value to the interpretation. Due to limitations on data availability in Central and Eastern Europe, the most suitable measurements and their respective advantages and disadvantages are listed below.
Employment
* Easy to understand, most readily available.
* Employers are more likely to give employment information than sales information.
* Useful in an urban planning context
* May inconsistently or inadequately measure part-time, seasonal, overtime work and differences in productivity and losses due to strikes.
* Is sometimes classified differently between countries and between cities and their own countries.
Payroll
* Good supplement to employment and sales data.
* Can be converted to wage rates for comparison purposes.
* Difficult to use without employment data.
* Difficult to obtain current information, and older information requires adjustment
Sales
* Most commonly available and most useful after employment data.
* Requires import or total market demand information to allocate between basic and nonbasic, which can be difficult to obtain below the national level.
* More complex and difficult to interpret than employment data.
* Confidentiality concerns may limit willingness to provide data.
Production Measures
* Number of units produced or tons of product shipped are useful measures for economic activities that produce physical outputs, but cannot be applied to all forms of economic activity.
* Value added is useful for understanding productivity and has greater universal applicability.
* Not universally available or applicable to all types of economic activity.
Classification of Economic Activity
To conduct an economic base analysis, the measure of economic activities must be classified into categories to allow comparison and measurement of change. In theory, the economic researcher must make decisions related to the level of aggregation versus detailed classification. In the reality of Central and Eastern Europe, the level of classification detail will be limited by the availability of data. The first level of aggregation is generally referred to as the "divisions" of economic activity. The next level of detail is referred to as the "major groups." In Western economies, Standard Industrial Classification systems will break down the major groups into two more levels of classification. The economic researcher must ensure that, when comparing international data, the data have been classified in the same manner as the local economy. The following chart shows a typical classification of divisions; a few major groups are indicated as examples of the next level of classification.
Typical Divisions of Economic Activity
Within urban areas, the first four divisions noted (agriculture through forestry) are not major economic activities, and are frequently aggregated together and called "primary activities."
The greater the level of detail with which the urban economic base analysis is conducted, the greater will be the understanding of the exact nature of the local economy under study. For example, if a city has a particular strength in retail trade but a weakness in wholesale trade, and if the economic researcher is only examining at the aggregate division level of "trade," neither the strength nor the weakness will be evident, as both will be masked by the aggregation of the two. This is particularly true in divisions with many major groups and a large proportion of employment, such as manufacturing.
In addition, if there is a particular division or major group that is indicated as a strength within the city, the economic researcher may wish to analyze to the next level of classification to learn more about that particular aspect of the local economy.
Caution must be used when comparing international data, as major groups may be classified under different divisions in different countries. For example, utilities are frequently classified under industry (the manufacturing division in Western economies) in Central and Eastern Europe; in Western economies, utilities are frequently classed under the transportation, communications, and utilities division.
Selecting the Methodological Approach
As noted in the preliminary considerations of economic base analysis, research procedures to describe and explain the economic base utilize either a direct or indirect approach. The primary goal of both approaches is the same—that is, to determine the extent of basic and nonbasic activity in each economic activity examined. A brief description and summary of the strengths and weaknesses of the two approaches follow.
Direct Approach
* The direct approach uses data gathered from local firms using a survey instrument such as a mail-in survey or personal or telephone interview.
* Properly executed, the direct approach can result in greater accuracy in identifying basic and nonbasic proportions of economic activities than can the indirect approach.
* Surveys must be properly designed, executed, and interpreted: this makes them expensive and time consuming; also, they require appropriately trained personnel.
* The accuracy of the results can be affected by a survey respondent's recollection of key data and familiarity/understanding of the geographical areas referred to in the survey.
* The survey instrument can be used to gather additional information of particular importance in understanding the city under study, such as linkages between firms.
* It is difficult to repeat this approach in a consistent fashion to develop time-series data.
* Patterns of stability or change cannot easily be described or explained until the approach is repeated at a different point in time.
Indirect Approach
* The indirect approach uses statistical data to compare the local economy with one or more "benchmarks" to estimate the degree to which individual economic activities are basic or nonbasic.
* The indirect approach produces relatively crude results that require thorough understanding of the analytical tool being used and its weaknesses and limitations, and very careful interpretation of results aided by a strong knowledge of the local economy and the benchmarks under study.
* This is a simpler, inexpensive application, requiring less specialized technical knowledge than the direct approach.
* With the indirect approach, it is easier to describe and explain patterns of stability or
change, as historical data are usually available for comparison purposes.
An example of a combination of direct and indirect approaches is included as appendix A to this document, "Vilnius Market Profile." The second part of this publication provides an in-depth examination of the current and potential export markets for manufacturing industries in Vilnius. It uses data from a number of sources verified by interviews. It does not examine exports from nonmanufacturing economic activities, nor does it examine imports due to a lack of import data.
Because of the complexity, time, and cost required, as well as the need for specialized expertise, the direct approach is not examined further, as it is beyond the reach of the intended users of this handbook. The indirect approach and its analytical tools, being easier and less expensive to use, is therefore recommended and described in further detail.
Determining Analytical Tools for the Indirect Approach
There are several techniques or analytical tools that are available to conduct economic base analysis. Some tools, however, are less reliable or less commonly used and are thus not reviewed here; these include the Course Sectoral Approach, the Index of Surplus Workers Method, and the Direct Ratio Method. The methods reviewed here are, from simplest to most complex, Location Quotient (LQ) Method; the Minimum Requirements Method, which supplements the LQ Method; and the Shift-Share Analysis. A brief description and summary of the strengths and weaknesses of each of these are presented below.
Location Quotient Method
Description
* Identifies basic and nonbasic activities for a city.
* Provides a measure of the concentration—also referred to as the "coefficient of specialization"—and importance of an economic activity in a city relative to the selected benchmark(s). Benchmarks may be other cities, the average of several other cities, the nation, or one or more international benchmarks.
* Based solidly on the economic base analysis method of analyzing local economies.
* Concept is:
_ If the city being studied has a higher concentration of an economic activity than the benchmark, this indicates a basic activity that "exports its surplus" (produces goods/services in a volume greater than required to meet the consumption needs of the local population).
_ If the concentration is less than the benchmark, the activity is nonbasic and the city is a net "importer" of that product or service (produces goods/services in a volume less than required to meet the consumption needs of the local population).
_ If the concentration is similar to the benchmark, the activity is nonbasic and the city is neither an exporter or importer, but is more or less "self-sufficient" in the provision of that product or service (produces goods/services in a volume that exactly meets the needs of the local population).
Advantages
* Inexpensive.
* Easy to do.
* Data availability is usually good.
* Has a relatively universal application, including for international economies.
* Historical data can be used to explain changes and trends in the economy.
* Produces conservative estimates of basic activities because the estimate is relative to the benchmark which will have a portion of basic activity already built in.
Disadvantages
* Assumes similar productivity, consumption, and "self-sufficient" product mix as the benchmark.
* Consistently understates the concentration of basic activities because the benchmark is assumed to be "self-sufficient"—that is, not an importer or exporter. In fact, all benchmarks will have some degree of basic activity.
* A larger geographic study area increases the understatement of the concentration of basic activity.
* The larger the aggregation of economic activities, the greater the chance that a subsector of basic activity will be hidden or masked by the presence of a nonbasic subsector—effectively canceling each other out. For example, if a city has a particular strength in retail trade but a weakness in wholesale trade, and if the economic researcher is only examining at the aggregate division level of trade, neither the strength nor the weakness will be evident as both will masked by the aggregation of the two. This is particularly true in divisions with many major groups and a large proportion of employment, such as manufacturing.
Minimum Requirements Method
Description
* Based on the premise that the degree of concentration of basic activity should be determined by comparison to the minimum percentage of an activity found in comparable cities (versus LQ, which compares against the average of the selected benchmark).
* Comparisons are made to a large number of similar cities.
Advantages
* Inexpensive.
* Easy to do.
* Has a relatively universal application, including for international economies.
Disadvantages
* Assumes that the minimum percentage found represents a "self-sufficient" city that is meeting local demand in the economic activity compared, when in fact the city is likely importing products or services in that economic activity.
* Adjustments usually need to be made to the minimum percentage found in each compared economic activity due to the above assumption, but any adjustments made are arbitrary. Without adjustments, basic activities could be overstated.
* There may be a limited number of similar cities with which to compare.
Shift-Share Analysis
Description
* This method is used by economic researchers to explain changes over time in an economy.
* It has the built-in capability of explaining changes over time by allocating changes in economic activity to three main sources of growth or decline.
_ National growth: the part of growth or decline attributable to growth, stability, or decline of all economic activities (totaled) that occurred nationally (or in the benchmark area chosen). This component is the overall growth in the benchmark economy.
_ Industry mix: the part of growth or decline attributable to growth/decline of the economic activity at the national (or benchmark) level (with the national growth component removed). This explains the growth of an economic activity that is not due to the overall growth in the benchmark economy.
_ Competitive shift: the part of growth or decline attributable to the growth/decline of economic activity at the local level (with the industry mix and national growth components removed). This explains the
growth of an economic activity at the local level that is not due to either the overall growth in the benchmark economy or to growth in the economic activity at the national (benchmark) level.
* The benchmark chosen is most frequently a national benchmark, but in strongly interrelated economies in Central and Eastern Europe, other benchmarks could reasonably be used, such as an average of the three Baltic states as a benchmark for a Baltic city.
Advantages
* Has greater descriptive power than the LQ or Minimum Requirements Methods.
* Inexpensive.
* Relatively easy to calculate.
* Data availability is usually good.
* Has a relatively universal application, even internationally.
Disadvantages
* Is more complex and difficult to interpret and explain.
* Only one of the three explanation components is based upon local changes; the other two are changes at the benchmark level.
* The three explanation components are not independent of each other; a change in one will affect the other two.
* Does not take into account changes in employment structure during the study period.
* Greater caution is required in interpreting results, and results should be checked for consistency and reasonableness against other sources.
* Requires time-series data to have been collected and compiled in a consistent fashion for both the area under study and the benchmark area.
* Does not reveal the degree of basic and nonbasic employment activity in the local economy, but rather describes the nature of
the changes that occurred in the local economy between two time periods.
* Comparisons cannot be made against unrelated economies, as the methodology requires the local economy to be part of the benchmark economy.
Selecting and Applying the Right Analytical Tool
The Shift-Share Method provides the greatest capability of describing and explaining changes in the local economic base (growth, decline, or stability) over time. Its application in Central and Eastern Europe presents several difficulties, however. All former Soviet republics have undergone massive structural changes in their economies in a very short period of time; these changes are not comprehended by this analytical tool. Different benchmarks would therefore produce different results, depending upon the nature and rapidity of change of the selected benchmark. Furthermore, this analysis requires that time-series data be collected and compiled in a consistent manner. Consistent time-series employment data are simply not available at the urban level—and whether national time-series employment data have been consistently collected and compiled is questionable. As well, international benchmarks can only be used within strongly
interrelated economies such as within the Baltics. This technique will, however, have greater applicability in the future, when these difficulties are no longer present.
The Location Quotient Method is therefore the preferred analytical tool for economic base analysis in Central and Eastern Europe. Because of the data collection and reliability issues in most transitional countries in the region, if at all possible the LQ Method should be supplemented by the Minimum Requirements Method to assist in interpretation and as a verification for reasonableness and consistency.
There are many data issues that could plague any analytical tool utilized, and data issues tend to arise more frequently and with greater seriousness in countries that have undergone massive changes such as the former Soviet republics. Extreme caution must therefore be utilized in the application of economic base analysis tools and in the interpretation of the results.
How to Apply the Location Quotient Method
The following box describes the steps involved in using the LQ Method. Refer to the previous section for issues to be considered in its application.
Applying the Location Quotient Method
1. Select the boundaries of the study area.
2. Choose the year (or years, if changes through time will be described) for which data are available.
3. Select one or more benchmark areas to compare to. In Central and Eastern European countries, this should include at minimum the country in which the city is located. Other useful benchmarks could be a competing city, the average of several nearby countries such as the Baltics for a Baltic city, and, optionally, perhaps a Western country for comparison purposes.
4. Decide on the level of aggregation that will be used depending on the availability of data for the city being studied and for the selected benchmark(s). Keep in mind that a greater disaggregation will produce greater accuracy of results.
5. For a selected economic activity, calculate its location quotient using the following formula.
Employment in economic activity e in city c
Total employment in city c
LQce =
Employment in economic activity e in benchmark area
Total employment in benchmark area
where: LQce = Location Quotient for economic activity e in city c
c = city under study
e = economic activity under study
6. Repeat the above calculation for each economic activity and place into a chart, or use a computer spreadsheet. (A sample spreadsheet with formulas follows.)
7. The result of each calculation will be a coefficient of specialization, or LQ, that will be<1, >1, or =1. The interpretation of the coefficients follows.
* If the value of LQce <1, then city c has less than its proportionate share of the economic activity e in comparison to the benchmark area, and city c is a net "importer" of activity e. This activity should be classified as nonbasic.
* If the value of LQce >1, then city c has more than its proportionate share of the economic activity e in comparison to the benchmark area, and city c is a net "exporter" of activity e. A portion of this activity should be classified as nonbasic, and the remainder as basic. If LQce = 1.5, then two-thirds of employment in this activity is nonbasic, and one-third can be classified as basic.
* If the value of LQce =1 or very close to 1, then city c has its proportionate share of the economic activity e in comparison to the benchmark area, and city c is neither an importer or exporter of activity e, but is more or less "self-sufficient" in the provision of that product or service. The activity should be classified as nonbasic, unless there is clear local knowledge that part of this activity
Spreadsheet Formulas for Location Quotients
How to Apply the Minimum Requirements Method
The following box describes the steps involved in the use of the Minimum Requirements Method. Refer to the previous section of this handbook for issues to be considered in its application.
Applying the Minimum Requirements Method
1. Select the boundaries of the study area.
2. Choose the year for which data are available.
3. Select a number of cities (perhaps 10 to 12) to compare that share similarities (usually population size).
4. Decide on the level of aggregation that will be used depending on the availability of data for the cities being compared. Keep in mind that a greater disaggregation will produce greater accuracy in the results.
5. For each city, calculate the percentage of total employment that is represented by the number of employees in each economic activity.
6. For each economic activity, rank the cities in order of increasing percentage of total employment.
7. For each economic activity, the smallest percentage of total employment is assumed to represent the "minimum requirement" to meet the needs of the local population, and this therefore nonbasic.
8. The economic researcher must then use his/her knowledge to determine the validity of the calculated minimum requirement percentages. The minimum found could be in a city that does not meet its local demand, but rather imports some, or a substantial part, of its needs. This effect can be exaggerated in smaller urban areas. The researcher determines if a percentage greater than the minimum should be used to obtain valid results.
9. The percentage of employment above the determined minimum requirement in each economic activity for the city being compared is therefore assumed to be basic.
10. Apply the determined minimum percentage to the total number of employees found in that economic sector in the city being studied to calculate the minimum or nonbasic employment.
How to Apply the Shift-Share Method
The following box describes the steps involved in the use of the Shift-Share Method. Refer tot he previous section of this handbook for issues to be considered in its application.
Applying the Shift-Share Method
1. Select the boundaries of the study area.
2. Select one or more benchmark areas to compare to. In small Central and Eastern European countries, this would normally be the country in which the city is located. In larger countries, if data are available on a regional or sub-country level, this data may be used as a benchmark. A completely separate analysis is conducted for each benchmark.
3. Choose the two time points for which data are available for the selected benchmark and the city. As shift-share measures change over time, consideration must be given to the number of years between data points. In rapidly changing economies such as in Central and Eastern Europe, no more than a five-year span between comparison points should be used; in slower changing economies, or to gain a longer term historical perspective, up to 10 or more years could be used between data points.
4. Decide on the level of aggregation that will be used, which will be somewhat dependent on the availability of data for the city being studied and the selected benchmark. Keep in mind that a greater disaggregation (more detailed breakdown) will produce more accurate results.
5. Determine the employment data for the first and second time points of each economic activity for the city under study. Repeat for the chosen benchmark.
6. The following formula is the basis for the shift-share analysis.
Aec = NGec + IMec + CSec
where: Aec = absolute change in employment between two time points for economic activity e in city c NGec = the national growth component of growth/decline for economic activity e in city c
IMec = the industry mix component of growth/decline for economic activity e in city c
CSec = the competitive shift component of growth/decline for economic activity e in city c
7. Calculate the values for each economic activity as follows:
NGec = E(T1) x R [Employment for Time Point 1 times the rate of growth for all economic activities in the benchmark area]
IMec = E(T1) x (Re - R) [Employment for Time Point 1 times (benchmark rate of growth for economic activity e minus the rate of growth for all economic activities in the benchmark area)]
CSec = E(T1) x (Rec - Re) [Employment for Time Point 1 times (rate of growth for economic activity e in city c
E(T1) = Employment in economic activity e for city c at Time Point 1
E(T2) = Employment in economic activity e for city c at Time Point 2
For each economic activity in city c, the absolute change in employment between T1 and T2 is allocated over the three components or sources of change as calculated in steps 6 and 7, which, again, is summarized by the formula: Aec = NGec + IMec + CSec
8. The resulting values can be charted and interpreted for each economic activity at the local level.
9. The values for each of the three components of national growth, industrial mix, and competitive shift can also be summed and their effects interpreted for the local economy as follows, assuming there was stability or growth generally in the local and benchmark economies. Local or benchmark economies that suffered employment declines between time periods require a careful interpretation, which is different than the one described below.
* High values for national growth indicate that growth of the local economy is due mainly to the overall growth in the benchmark economy.
* High values for industrial mix indicate that the local area has a high proportion of economic activities that experienced growth at the benchmark level. Low values indicate an absence of economic activities that experienced growth at the benchmark level.
Economic Base Analyses of the Baltic Capitals
Preparing the Economic Base Analysis for the Baltic Capitals
The economic base analyses prepared in this handbook are the first that have been prepared in the Baltic states. Time-series data were not available for any of the three Baltic capitals, so a shift-share analysis could not be performed. The preferred tool for economic base analysis in transition economies, the LQ Method, was used. Location quotients were calculated for each Baltic capital compared against its own country and against the average of the three Baltic countries together (where data were available). As the economies are closely related—and, to a certain extent, the capitals are competitors—it is useful to compare each city's LQs against the Baltic averages, thereby deriving a deeper understanding of each local economy.
Several problems were encountered in preparing the economic base analyses.
* Employment data at the city level is very difficult to obtain and has limited reliability.
* Employment data may be classified differently at the national and local levels within a country.
* Data are classified differently from country to country. For example, in Estonia, financial and insurance employment is included in service sector employment.
* Tallinn and Estonia employment data appeared to be particularly weak, with a higher level of aggregation. No disaggregation of manufacturing employment, for example, is available.
* Time-series employment data were not available at the city level.
* There is very limited experience and varying degrees of expertise available in the capi
tal cities to interpret the results of the economic base analysis.
As noted at the beginning of part IV, the ability of these procedures to explain the local economy and its relationships is highly dependent on two things: the quality of the information available and the quality of the local knowledge that must be used to interpret and verify the results of any quantitative method. And, as noted above, there are significant problems related to both of these issues. The results of these research procedures should thus only be used as an example of how to conduct and interpret an economic base analysis. They must not be relied on in their current form for the development of economic development programs.
To be able to use the results of economic base analyses, the following would be required:
* All of the data would require a complete verification.
* The location quotients would need to be recalculated with verified or adjusted data.
* The analysis should be backed up with other methods such as the Minimum Requirements Method.
* Other data measurements should be used for validation purposes (sales, exports, imports, production measures, etc.).
* Significant review and study to interpret the results should be undertaken by experts with the best knowledge of the local economy.
Riga Economic Base Analysis
For this study, the base analysis looks at the distribution of employment in the Riga region and compares it against both Latvia and the Baltics.
Based on the 1996 employment data and location quotients presented in the following table, the transportation-communication and financial-insurance employment sectors produced the equally large location quotients of 1.88. The concentration of financial and insurance em
ployment in Riga is not surprising given that Riga is a capital city, and the capital is frequently the financial center of the country. For this reason, insurance and other business services also tend to concentrate at the financial center. Riga's LQ of 2.01 for financial-insurance and 2.11 for real estate are also the highest LQs compared to the Baltics. This could be explained by the fact that Riga is the largest capital city, is the dominant city in Latvia, and has developed most rapidly of the Baltic capitals. The high concentration of real estate_related employment in Riga results from Riga's rapid development and the fact that, as a business service, real estate activity tends to concentrate at the country's financial center.
The large LQ (1.88) for transportation-communication is partially explained by the importance of Riga's port within Latvia and employment related to this port and associated activities. The LQ for Riga's transportation-communication sector compared against the Baltics (1.95) indicates a similar strength. This again could be explained by the fact that while transportation and communication employment represents 8 to 10 percent of all employment in the three Baltic countries, Riga's employment in transportation and communication represents 17 percent of Riga's employment.
Public administration has a relatively high LQ of 1.53—not surprising, given that Riga is the capital city, the seat of the national government. In addition, with more than 40 percent of Latvia's population in the Riga region, one could expect relatively high levels of local government administration to contribute to this high LQ. Within the Baltics, Riga has a similar concentration of public administration with an LQ of 1.52.
Latvia's manufacturing is somewhat concentrated in Riga (LQ of 1.29), but its LQ compared against the Baltics is somewhat less at 1.06. Within the manufacturing sector, Riga appears to have a concentration within in several
subsectors: leather articles (2.59), utilities (1.90), clothing and furs (1.76), furniture manufacturing (1.67), and machinery and equipment (1.64). These Latvian manufacturing subsectors are concentrated in Riga. The manufacturing industries of food products, textiles, pulp and paper, and chemicals show no particular concentration in Riga, or are slightly underrepresented.
Education (0.95), health and social work (0.94), services and other employment (0.96), and trade (0.90) are not that different from national shares; therefore Riga is not at any competitive advantage in these employment sectors either in Latvia or within the Baltics. This does not suggest that Riga does not have strength in the education levels of its workforce, only that education as an employment activity is not necessarily concentrated in Riga.
Riga Location Quotient Analysis
Construction employment in Riga, as with most urban areas, is underrepresented compared with both Latvia (0.74) and the Baltics (0.70); this suggests that this somewhat land-extensive employment sector, which is also sensitive to land costs/taxation, is located outside the urban area
of Riga. Of course, much of Latvia's construction activity actually takes place in Riga, but the companies undertaking the construction are likely to be located outside Riga.
Employment in the primary sector (mining, agriculture, fishing, hunting) is severely underrepresented in Riga compared to both Latvia (0.06) and the Baltics (0.09); this is not at all surprising, as these employment activities are nonurban in nature.
The basic to nonbasic ratio suggested by the employment figures is 1.00:3.59, meaning that for every job created in basic economic activities, there are 3.59 jobs created in nonbasic activities. This ratio must not be used, however, until the employment figures have been confirmed and the LQ analysis and this ratio recalculated. (Also see the discussion of Local Economic Development/Economic Forecasting at the end of this part.)
Tallinn Economic Base Analysis
For this study, the economic base analysis looks at the distribution of employment in the Tallinn
region (1997), and compares it against both Estonia and the Baltics (1996). While it is not ideal to compare data from different years, these were the only employment data available at the time of analysis. Unfortunately, employment data for manufacturing subsectors were not available; therefore, no analysis can be performed on the subsectors. It should be noted that employment data accuracy is particularly questionable for Tallinn, and must not be relied upon in any way.
Based on the 1996-97 employment data and location quotients presented in the following table, the service sector and other services has the highest LQ of 1.41 compared against Estonia, and an even higher LQ of 1.53 compared against the Baltics. Service sector growth has been quite strong in Tallinn, which is at least partially supported by strong tourism, especially from Finland. Also artificially inflating this LQ is the fact that all financial and insurance_related employment is included in the service sector employment figure. The higher Baltic LQ results from a relatively (in the Baltics) low level of employment in the service sector in Latvia. Tourism may also underlie the high LQ of 1.29
(Estonia) and 1.97 (Baltics) for hotel and restaurant employment, which is part of the service sector.
The large LQ (1.39) for transportation-communication is explained by the importance of Tallinn's ports within Estonia and employment related to the ports and associated activities. Tallinn's ports handle a high volume of goods and a high number of tourists, especially from Finland. In addition, the Estonian telephone company and most of the cellular telephone companies are headquartered in Tallinn. The LQ for Tallinn's transportation-communication sector against the Baltics (1.57) indicates even greater strength, which could be explained by the fact that Lithuania has proportionately less employment in transportation and communications, which would therefore raise Tallinn's LQ as compared against the Baltics. Latvia and Estonia have approximately the same percentages of employment in this sector (10 percent).
The wholesale and retail trade sector is also somewhat concentrated in Tallinn with an LQ of 1.18 versus Estonia. Generally speaking, retail trade has been growing very quickly in the
past two years in Tallinn, with a 20 percent growth in the number of retail stores in 1996 alone. This growth has also been supported by tourism to Tallinn. The higher LQ for this sector versus the Baltics (1.32) results from the relative underdevelopment of the wholesale and retail sector in Lithuania as compared to Estonia and Latvia. This is supported by the fact that wholesale and retail trade represents only 8.4 percent of all Lithuanian employment versus 14.2 and 13.3 percent in Latvia and Estonia, respectively. Tallinn's employment in this sector is 15.7 percent of all Tallinn's employment.
For a city, Tallinn has a relatively high concentration of construction employment with an LQ of 1.34 compared to Estonia, and 1.22 compared to the Baltics. Normally, construction companies require more extensive land use, frequently locating outside a city to take advantage of outside storage of materials on relatively inexpensive land. Tallinn's high LQ can be explained by the presence of several large construction firms (as well as many smaller companies) headquartered in Tallinn (where the employment will be counted). Many of these firms have warehouses and other facilities outside the city.
Education, health and social work, and public administration do not have any particular concentration of employment, with LQs of 0.93, 0.87, and 0.94 as compared against Estonia. The LQs indicate underdevelopment of these sectors in comparison with the other Baltic countries—0.77, 0.66, and 0.88, respectively. The LQ for health care and education is partially explained by the fact that Tartu, Estonia's second largest city, is a strong university and research center.
Employment data are not available for the primary employment sector in Tallinn.
A basic to nonbasic ratio was not calculated due to questionable data accuracy.
Tallinn Location Quotient Analysis
Vilnius Economic Base Analysis
For this study, the base analysis looks at the distribution of employment in the Vilnius region and compares it against both Lithuania and the Baltics.
Based on the 1996 employment data and location quotients presented in the following table, the medical, precision, and optical instruments manufacturing industries have a very high LQ of 3.29. While this is not Vilnius's largest manufacturing industry, this Lithuanian industry is highly concentrated in the city, and approximately 60 percent of its production was exported outside of Lithuania in 1996. Most companies are small, employing fewer than 20, and are involved in scientific and high technology manufacturing. A sampling of the companies follow:
* Brown Sharpe-Precizika and Matas, producing measuring instruments and appliances;
* Intersurgical, producing surgical instruments;
* Isra, producing portable projection devices;
* Light Conversion, producing laser parametric generators;
* Rimeda, producing medical equipment;
* Soveta-Baltica, producing disposable medical appliances; and
* Vitlita, producing medical equipment related to plasmapheresis.
The utilities sector produced the next largest location quotients of 3.06 compared against Lithuania, and a very high 4.06 against the Baltic states. The concentration of utility employment can be partially explained by the inclusion of all employees of two large state utilities (and their subsidiaries)—Lithuanian Gas and Lithuanian Energy—within the Vilnius employment figures. In fact, many of these employees are not in Vilnius, but are included in the statistics because the head office is in Vilnius. Vilnius Water is also a large employer. Vilnius's even higher LQ of 4.06 compared against the Baltics might be explained by the privatization and pos
sible decentralization of utility companies in Latvia and Estonia.
Vilnius has a high concentration of Lithuania's employment in rubber and plastic product manufacturing (LQ of 2.72). This is explained by the fact that more than 99 percent of all plastic film and tubes in Lithuania is produced in Vilnius, mainly by one of Vilnius's largest enterprises, Plaster.
The printing and publishing industry is also strong in Vilnius with an LQ of 2.24, accounted for by the following:
* The main Lithuanian daily newspapers have their headquarters and printing plants in Vilnius.
* The majority of Lithuania's largest publishers are in Vilnius.
* The largest companies engaged in printing and publishing (business directories, catalogues, etc.) are in Vilnius.
Vilnius Location Quotient Analysis
Vilnius has a concentration in the manufacture of machinery and equipment (LQ of 1.83), with 20 percent of total manufacturing sales in 1996. This is explained by the presence of 3 of the 10 largest manufacturing firms in Vilnius being manufacturers in this sector.
Pulp and paper industries also have a strong presence in Vilnius (1.36), with 5.7 percent of total manufacturing sales in 1996. These industries include the manufacture of paper and paper articles, packaging, and cardboard.
The concentration of financial and insurance employment in Vilnius (LQ of 2.09 as compared against Lithuania) is not surprising given that Vilnius is a capital city, and the capital city is frequently the financial center of the country. For this reason, insurance and other business services also tend to concentrate at the financial center. All the headquarters of Lithuania's largest banks are located in Vilnius. The headquarters of most insurance companies are in
Vilnius as well. Vilnius's LQ of 2.25 for financial and insurance is the second highest LQ against the Baltics. This could be explained by the fact that the three capital cities are all the dominant financial centers within their countries; therefore all three will show high LQs against Baltic employment in this sector.
The high concentration of business services (LQ of 2.15) and real estate_related employment (LQ of 1.67) in Vilnius results from the fact that business services, including real estate activity, tend to concentrate at the financial center of the country. These high LQs suggest that Vilnius is serving as the business service and real estate center for Lithuania. Vilnius's LQ of 0.46 for real estate compared against the Baltics would indicate that this Lithuanian economic activity is still underdeveloped in relation to its Baltic neighbors. Real estate employment in Lithuania represents less than 1 percent of all employment versus 5 percent in both Latvia and Estonia.
The LQ of 1.37 against Lithuania for transportation and communications is an artificially high LQ. Similar to the utility sector, the presence in Vilnius of the head offices of two large enterprises results in all employees (including subsidiaries) located outside of Vilnius being included in Vilnius employment figures. Lithuanian Telecom and Lithuanian Railway are both located in Vilnius. The LQ for Vilnius's transportation-communication sector against the Baltics (1.19, which would also be artificially inflated) does not indicate any particular strength; this could be explained by the fact that each of the three Baltic countries has significant port-related transportation employment (Vilnius is not a port city).
The wholesale and retail trade sector is strong in Vilnius with an LQ of 1.28 against Lithuania. Generally speaking, retail trade has been slower to develop in Lithuania than in the other Baltic states, but in the last three to four years, retailing has boomed in Vilnius. With only 16 percent of the country's population, Vilnius had 25
percent of retail trade turnover in 1995, with no slowdown in sales growth evident. Vilnius is also a major wholesale trade center, with 10 of Lithuania's largest wholesale trading companies, as measured by turnover. The wholesale companies are engaged in the sale of computers, software, copying equipment, and cars. Vilnius's LQ in wholesale/retail trade against the Baltics is 0.91, which indicates that Vilnius does not have a particular strength in this sector within the Baltics. The spread between these two numbers would tend to indicate that wholesale and retail trade is generally underdeveloped in Lithuania, and average for Vilnius in the broader Baltic context. This is supported by the fact that wholesale and retail trade represents only 7.1 percent of all Lithuanian employment versus 14.2 and 13.3 percent in Latvia and Estonia, respectively. Vilnius's employment in this sector is 8.4 percent of all Vilnius employment. Another important factor to note is that Vilnius and Lithuanian trade employment figures do not include the self-employed, many of whom work in wholesale and retail trade.
Public administration shows only some mild strength with an LQ of 1.19; this is not surprising, given that Vilnius is the capital city, the seat of the national government. Within the Baltics, Vilnius has a similar concentration of public administration with an LQ of 1.26.
The manufacture of clothing and furs (1.16) and leather articles (1.15) in Vilnius shows only a mild concentration within Lithuania, but a more marked concentration within the Baltics (1.57 and 1.40, respectively). This is explained by the fact that both industries have a greater concentration of national employment in Lithuania than the same industries within Latvia. The manufacturing employment in electronics and electronic technical equipment is about average in Lithuania (0.94).
Education (0.81) and health and social work (0.82) are slightly lower than national shares; therefore, Vilnius is at a slight competitive dis
advantage in these employment sectors in Lithuania, but average within the Baltics (1.0 and 1.05, respectively). This might be explained by the fact that Vilnius has a smaller percentage of Lithuania's population (and therefore a smaller percentage of employees), in comparison to the other Baltic capitals which are the dominant cities in their countries.
Construction employment in Vilnius is underrepresented compared against both Lithuania (0.74) and the Baltics (0.83). This sector is a somewhat land-extensive employment sector, and one which is also sensitive to land costs/taxation; it therefore may be located outside of the urban area of Vilnius.
Employment in the primary sector (mining, agriculture, fishing, hunting) is severely underrepresented in Vilnius as compared with both Lithuania (0.02) and the Baltics (0.01); this is not at all surprising, as these employment activities are nonurban in nature. Manufacturing industries underrepresented in Vilnius include food products (0.43), textiles (0.29), furniture (0.83), construction materials and other nonmetallic products (0.49), basic and fabricated metals (0.60), wood and wood articles (0.47), and chemicals/chemical products (0.59).
The basic to nonbasic ratio suggested by the employment figures is 1.00:3.09, meaning that for every job created in basic economic activities, there are 3.09 jobs created in nonbasic activities. This ratio must not be used, however, until the employment figures have been confirmed and the LQ analysis and this ratio recalculated. (Also see the discussion of Local Economic Development/Economic Forecasting at the end of this part.)
Using the Results of an Economic Base Analysis in Economic Development
The power of conducting an economic base analysis is only realized if its results and interpretations can be utilized in a city's local eco
nomic development program. The results of the economic base analysis merely suggest directions that should be examined in further detail. A city should not act on the analysis results alone, but must first validate the results and then use them as a pointer to where it should further investigate economic development actions or strategies that might be utilized.
The following ideas are provided as examples only of the type of investigations that could be suggested by the economic base analyses conducted for the three capital cities. Note that these are examples only, because an economic base analysis must be based on verified data and appropriate interpretation. These examples must not be relied upon in any way.
* The high Riga LQs for finance, insurance, and real estate suggest further investigation of why the employment figures are high. Interviews with industry representatives and other research may confirm that this may be an area of strength upon which Riga could build as a "center of expertise" or, in other words, an "engine of growth."
* Further investigation is also required into Riga's high transportation-communication LQ. Does this result from employment concentration in one or both of these sectors? It is possible that Riga could further build on this strength as well.
* Several of Riga's manufacturing subsectors show strength that should be investigated further to determine if there are opportunities to begin building "clusters" of like industries.
* In Tallinn, the service sector warrants much closer investigation to determine the causes of its high LQs. Tourism is a contributor here, but there may be other reasons as well. The strength suggests a sector around which Tallinn may wish to build part of its economic development strategy.
* Tallinn's high LQ in transportation and communications requires further investigation to determine if one or both sectors are causing
the high LQ. Whatever it is suggests again that some economic strategy is warranted to build on this strength.
* Vilnius has a very high LQ for the manufacture of medical, precision, and optical instruments. Upon confirmation, a strategy is suggested that may try to build upon this strength to create a cluster of related activity, supported by appropriate education programs, research and development, etc.
* The high Vilnius LQ for finance and insurance suggests further investigation into why the employment figures are high. Interviews with industry representatives and other research may confirm that this may be an area of strength that Vilnius could build upon as a center of expertise.
* Investigation is required to determine how much the Vilnius utility LQ is inflated by counting all Lithuanian employees associated with Vilnius head offices as Vilnius employees.
* Much more work needs to be done to understand wholesale and retail trade in both Lithuania and Vilnius. An underdevelopment in either of these might suggest that there are opportunities here.
* Several Vilnius manufacturing industries suggest economic development strategies. A prime example appears to be the related industries of printing and publishing and the manufacture of paper, packaging, and cardboard. A deeper understanding of the nature of the industries present may suggest how a clustering effect could be built into an engine of growth.
* Low LQs in any of the cities for economic activities that are typically served from within the city suggest opportunities for local entrepreneurs to bring their own product/service to market to "substitute" for imports into the city.
This is not an exhaustive list, but serves to illustrate where these cities might concentrate some further investigation efforts after verification of the economic base analysis.
Conclusions: Economic Base Analysis for the Baltic Capitals
The economic base analysis completed above demonstrates how such an analysis can help a city understand its local economy. As noted at the beginning of part IV, the ability of economic base analysis to explain the local economy and its relationships is highly dependent upon two things: the quality of the information available and the quality of the local knowledge that must be used to interpret and verify the results of any quantitative method.
Unfortunately, data availability and data consistency (collection and classification)—especially for time-series data—are serious concerns. Hopefully, it is evident through this economic base analysis demonstration how important accurate information is to completion of an accurate analysis. Again, due to concerns about the employment data, the results of the economic base analysis for the Baltic capitals should be regarded as an example only and should not be used in its current form.
To be able to use the results of the economic base analyses, the following would be required:
* All of the data would require a complete verification.
* The location quotients would need to be recalculated with verified or adjusted data.
* The analysis should be backed up with other methods such as the Minimum Requirements Method.
* Other data measurements should be used for validation purposes (sales, exports, imports, production measures, etc.).
* Significant review and study to interpret the results should be undertaken by experts with the best knowledge of the local economy.
Other Approaches to Explaining the Local Economy
There are other approaches available to describe and explain the local economy. All use a direct approach, meaning that field data must be collected and analyzed, unlike the indirect approaches described here which use statistical historical data. These approaches also require much more sophisticated and comprehensive resources to design, apply, and interpret them. A few of these methods are briefly described below.
Input-Output Analysis
Input-output (IO) analyses are survey-based studies that map the interdependencies among economic activities in a local economy through extensive surveys of local firms. The surveys focus on the magnitude and source of inputs required to produce units of output (goods or services). By understanding the source of inputs and the distribution of the outputs to firms and sectors, one can develop a comprehensive understanding of the local economy. The IO study, properly carried out, is the most comprehensive approach to understanding local economies, and provides a superior base of information for impact and forecasting studies.
Intersectoral Flow Method
This method is also survey-based and focuses on aggregate relationships between economic activity groups within the study area and on private and government exports. It identifies basic and nonbasic employment and direct and indirect multipliers. Surveys of sales data identify the flow of goods and services between economic activities and the sectors served; this information is then converted to employment data and allocated across all economic activities. Direct and indirect linkages are estimated based upon the survey data and basic to nonbasic ratios are calculated using these direct and indirect linkages. This method signifi
cantly improves on the results obtained by indirect approaches.
Cluster Analysis
The cluster, or industrial complex, analysis focuses only on a certain segment of a local economy, rather than on the whole economy. It is also survey-based, and reveals interdependencies of products, services, and processes within a geographical cluster of strongly interrelated firms in similar businesses. The objective is to determine locational activity and relationship patterns, which are then used to identify economic development opportunities. Cluster analysis is frequently used to study concentrations of high-technology firms in a city.
Local Economic Development/Employment Forecasting
Local employment forecasting, while very desirable, is an activity that is very difficult to do with a reasonable degree of accuracy. Most employment projection tools are made to be utilized on a macroeconomic scale and lose a significant degree of reliability when used on a microeconomic scale. Furthermore, all projection tools rely upon a clear and accurate description of the past behaviors of an economy over time. The projections then depend upon the assumption that the future behavior of the economy will roughly mimic the behavior of the economy in the past.
There are therefore three reasons that make local employment forecasting in Central and Eastern European countries very difficult and unreliable at this time:
* The tools available for local employment forecasting are inherently weak.
* Data available on the past performance of national economies in Central and Eastern Europe has a relatively high degree of unreliability in terms of accuracy and consistency of collection and classification, while data (regardless of reliability) at the level of
the local economy is almost nonexistent, especially for time-series data.
* Most economies in Central and Eastern Europe are still undergoing massive economic restructuring, so that the past performance and behaviors of these economies could not be used to predict their future behavior even if the data were plentiful and reliable and the projection tools were stronger.
As a result, until employment data are collected and classified in a consistent manner at both the national and local levels for several consecutive years, local employment forecasting will not be possible.
In the meantime, the most useful tool available to municipalities will be the basic to nonbasic employment ratio. The ratio can be used to predict the effects of economic development efforts to increase basic employment. Again, the ratios calculated in this handbook must not be used in their current form due to data reliability
problems. An example of the use of the ratio or "multiplier" follows.
In Vilnius, the calculated basic to nonbasic ratio was 1:3. In the medical, precision, and optical equipment manufacturing industry, there are 1,712 employees classified as "basic" and 746 classified as "nonbasic." Assuming local demand is constant and satisfied with the manufactured goods made by the 746 employees, an increase in employment in this industry would result in an increase in nonbasic employment through the multiplier effect. For example, an increase of 100 employees in this manufacturing industry would result in the creation of an additional 300 jobs in nonbasic economic activities (using the ratio of 1:3), for a total of 400 new jobs.
Again, this is an example of the use of the basic/nonbasic ratio only, and must not be relied on until the Location Quotient analysis is updated and verified as noted in the Conclusions section.
Bibliography
Bryant, Christopher R., and Richard E. Preston. 1988. Economic Development Bulletin Number 3: Information for Local Economic Development. University of Waterloo.
Bryant, Christopher R., Richard E. Preston, and Douglas J. Dudycha. 1988. Economic Development Bulletin Number 5: Marketing and Local Economic Development. University of Waterloo.
———. 1989. Economic Development Bulletin Number 7: Describing the Economic Base of the Local Economy. University of Waterloo.
———. 1990. Economic Development Bulletin Number 8: Explaining the Economic Base of the Local Economy. University of Waterloo.
McSweeney, Eric. 1997. CUI Guide to Economic Development and Strategic Economic Planning. Canadian Urban Institute.
Tiebout, Charles M., 1962. The Community Economic Base Study. Supplementary Paper No. 16. New York: The Committee for Economic Development.