Employment Change in LDs of West Virginia: A Dynamic Spatial Shift- Share Analysis
1Tillman School of Business, Mount Olive College/University, North Carolina, USA
2Davis College of Agriculture, Natural Resources and Design, Division of Resource Management, West Virginia University, USA
This paper uses dynamic spatial shift-share analysis to examine employment change in local development districts (LDs) of West Virginia for the period of 1976 to 2007. This paper differs from the standard shift-share analysis by combining the dynamic shift-share with the spatial shift-share and decomposing the employment change of selected sectors into four effects: state growth effect, industry mix effect, neighborhood competitive effect, and state competitive effect. Distance between main cities of local development districts were considered in developing the distance based weight matrix for measuring neighborhood competitive effects. Results describe the changing pattern of employment of the ten selected sectors namely farm employment, mining, construction, manufacturing, wholesale and retail trade, transportation and public utilities, finance insurance and real estate, services, government, and agricultural services of the eleven local development districts of West Virginia. In many LDs’ service, finance, insurance and real estate, construction, and government sectors have been the leading contributors of employment growth in 32 year period. Even though, mining and manufacturing are two main sectors in the economy of West Virginia, employment growth of mining is reported only in a few LDs. Manufacturing employment is declining in all LDs. All LDs have some sectors that are adding employees. Investing them might accelerate economic growth. Neighborhood competitive effects show potential spatial impacts on several sectors.
At a glance: Figures
Keywords: dynamic, spatial shift-share analysis, local development districts, West Virginia
American Journal of Rural Development, 2013 1 (5),
Received August 15, 2013; Revised September 05, 2013; Accepted September 25, 2013Copyright: © 2013 Science and Education Publishing. All Rights Reserved.
Cite this article:
- Herath, Janaranjana, Peter Schaeffer, and Tesfa Gebremedhin. "Employment Change in LDs of West Virginia: A Dynamic Spatial Shift- Share Analysis." American Journal of Rural Development 1.5 (2013): 99-105.
- Herath, J. , Schaeffer, P. , & Gebremedhin, T. (2013). Employment Change in LDs of West Virginia: A Dynamic Spatial Shift- Share Analysis. American Journal of Rural Development, 1(5), 99-105.
- Herath, Janaranjana, Peter Schaeffer, and Tesfa Gebremedhin. "Employment Change in LDs of West Virginia: A Dynamic Spatial Shift- Share Analysis." American Journal of Rural Development 1, no. 5 (2013): 99-105.
|Import into BibTeX||Import into EndNote||Import into RefMan||Import into RefWorks|
A regional economy consists of various sectors with a variety of economic potentials. A change in any of these sectors directly or indirectly affects the overall growth of the economy  and understanding the comparative advantage of these sectors becomes important in the development of decision making of the region [2, 3, 4, 5]. The identification of comparative advantaged sectors becomes more important in less-developed regions like West Virginia than in the developed regions.
West Virginia is a poor state in the Appalachian and southeastern regions. The total population of West Virginia is about 1.85 million and nearly 56 percent live in urban areas while the rest are in rural areas . West Virginia reported the 5th highest poverty rate, 17.4 percent during the year 2008  and at present, the rural poverty rate is 19.9 percent, a figure five points higher than the urban poverty rate. The unemployment rate is 9.5 percent, which is one of the highest rates compared to other states . The average population density is about 75 people per square mile, and the per capita personal income was $32,219 in 2009, 17.7 percent below the national per capita income of $39,138 . The per capita income in the rural area was $29,200 compared to $33,578 in the urban area in 2008 . Manufacturing, services, mining and tourism (an emerging sector) are the major economic sectors in West Virginia. The production of bituminous coal, natural gas, stone, cement, salt and oil contribute significantly to the state’s economy. Farming is practiced throughout West Virginia, but not in a form of large scale cash-crop agriculture . Nearly 82 percent of total agricultural production comes from livestock with the remaining 18 percent from crops. Major manufacturing goods in West Virginia include machinery, plastic and hard wood products, fabricated metals, chemicals, aluminum, automotive parts and steel. The largest share of gross product in the service sector is given by the community, business and personal services groups and most of these services are provided by private health care, law firms, repair shops and hotels. Tourist activities such as skiing and white-water rafting are popular and bring significant revenue to West Virginia.
According to the economic outlook of 2010 and 2011 , economic gains in West Virginia are slow and moderate compared to the national rates. To enhance economic growth, an understanding of the prevailing employment pattern within potential regions and industries is essential. Thus, the main objective of this paper is to analyze the employment changes in local development districts (LDs) of West Virginia as compared to the employment change in the state.
2. Local Development Districts of West Virginia
West Virginia is divided into eleven local development districts (LDs) each with a regional development council to provide comprehensive planning of all types for the state (Figure 1). Local development district 1 (LD1) includes the counties of McDowell, Mercer, Monroe, Raleigh, Summers, and Wyoming while Beckley in Raleigh County and Princeton, Bluefield, in Mercer county are the main cities in the region. LD2 includes Cabell, Lincoln, Logan, Mason, Mingo, and Wayne counties and Huntington in Cabell County is the main city of the district. LD3 consists of Boone, Clay, Kanawha, and Putnam County, and Charleston in Kanawha County is the main city. LD4 includes Fayette, Greenbrier, Nicholas, Pocahontas, and Webster County. The main city is Oak Hill in Fayette County. Calhoun, Jackson, Pleasants, Ritchie, Roane, Tyler, Wirt, and Wood County are included in LD5. Parkersburg in Wood counties is the main city of LD5. There are six counties in LD6, Doddridge, Harrison, Marion, Monongalia, Preston, and Taylor. Morgantown and Clarksburg are the main cities in the district. LD7 consists of Barbour, Braxton, Gilmer, Lewis, Randolph, Tucker, and Upshur. Elkins in Randolph County is the main city. LD8 includes five counties of Grant, Hampshire, Hardy, Mineral, and Pendleton County. Keyser in Mineral County is the small city in LD8. LD9 consists of Berkeley, Jefferson, and Morgan County. Shepherdstown and Kabletown in Jefferson and Martinsburg in Berkeley are the main cities in LD9. LD10 includes Marshall, Ohio, and Wetzel County. Wheeling is the main city. LD11 includes only two counties, Brooke, and Hancock. Weirton in Hancock County is the main city.
The paper is organized into six sections. Section 3 provides the literature review. Section 4 covers the methodology and data sources. Section 5 describes the empirical results and analysis. Section 6 presents the conclusions and policy implications.
3. Literature Review
The shift share method of analyzing regional growth was originated in 1940s by Daniel Creamer and was described as an analytical technique for understanding the regional development of a national economy by Dunn in 1960 . According to Dunn , the main feature of shift share analysis is the computation of geographical shifts in economic activity. The analysis has been used extensively since its formal inception in the fields of regional economy, political economy, urban studies, geography and marketing in the last 40 to 50 years [12, 13, 14, 15, 16, 17]. The technique is generally used to describe historical growth trends, forecast regional growth, analyze the effects of policy initiatives, or develop strategic planning for communities .
With Isard’s  finding that any spatial unit is affected by the positive and negative effects transmitted from its neighboring regions, numerous extensions and modifications were attempted to make up the drawbacks of standard shift-share analyses. Cliff and Ord [20, 21, 22] studied the problem of spatial autocorrelation. Hewings  included the spatial interaction in the shift-share model. Nazara and Hewings  and Mayor and Lopez  considered interregional interaction in the decomposition analysis by incorporating a spatial structure within shift-share analysis. Fernandez and Menendez  used spatial shift share analysis in measuring employment growth in Spain. The technique was applied to tourism in China, based on international tourism receipts from 1995 to 2004 . The study aimed at probing into the spatial competitiveness of international tourism in Jiangsu Province in comparison with its neighbors.
In 1988, Barff and Knight III developed the idea of dynamic shift-share analysis . While Harris et al. used dynamic shift share analysis to investigate the economic impacts of Nevada , Markusen et al. used dynamic shift share analysis in tracking the sensitivity of regional growth to international flows, decomposing shift-share components into import, export, and domestic market segments and a productivity component . Wilson et al. used dynamic shift share analysis to examine the export performances of China in electronics compared to the East Asian Newly Industrialized Economies exporting to the U.S, European Union and Japan between 1988 and 2001 . Herath et al. used dynamic shift share analysis in measuring economic growth in West Virginia .
In the context of regional economic growth analysis, various methods including optimization technique, shift-share, economic base, input-output, and benefit-cost analysis, have been used [18, 33]. Even though these models show some variations in measurement, precision, accuracy, and simplicity, each method of analysis has been used as a guide for policy makers and researchers in answering basic questions related to state or regional problems . Shift share analysis, in particular, compared to the other methods, is widely used by regional development practitioners, where data limitations are minimal [4, 33]. Furthermore, shift share analysis is important in selecting and understanding the key leading sectors in a region which needs developing local industry partnership.
4. Methodology4.1. Standard Shift-Share Model
The standard shift-share model decomposes a region’s employment growth into three effects as state growth effect (SGE), industrial mix effect (IME) and competitive effect (CE). NGE which is the part of change in total employment in a region ascribed to the rate of growth of employment at the state level, IME is the amount of change the region would have experienced if each of its industrial sector had grown at the state rates, less the SGE and CE is the difference between the actual change in employment and the employment change expected if each industrial sector grows at the state rate. The sum of these three effects gives the actual change in total employment within a region over a considered time period.
Assume eij, r, ri, and rij represent the employment of industry i in region j, the growth rate of state employment, the employment growth rate of industry i in the state, and the employment growth rate of industry i in region j. Then, the employment change (Δeij ) could be written as follows (equation1: where eij’ is the employment for the end year of the considered time period and eij is the beginning year of the considered time period.
The SGE is shown by eij* r ; IME is shown by eij*(ri – r), CE is shown by eij*(rij – ri). The total employment change can be determined by adding the three effects.4.2. Dynamic Shift Share Model
The dynamic shift-share analysis calculates the three effects annually and then summarizes the results over the study time period . Unlike the standard shift share analysis which considers only the origin and end years of a time period, this method considers all the annual changes of employment. Thus, the weak assumption of no or minimum employment changes within the period of origin and end year could be overcome by dynamic approach. The employment change (Δeij) can be shown as follows. As in the standard shift share analysis, eijt-1* rt shows the SGE, eijt-1 * (rit – rt) shows the IME, and eijt-1 *(rijt – rit) indicates the CE.
Standard shift-share model assumes that regions are independent from each other and there are no any positive or negative effects from neighboring regions. This assumption is weak and can be overcome with spatial shift share models where neighboring impacts on employment changes can be estimated. Thus, the results of a spatial shift share model are more accurate and useful than a standard shift share analysis.
The districts are spatial sub-units within a country or a state and the performances of a particular district may be affected by the performance of surrounding districts. Such influences can be captured using a spatial structure within the standard shift share analysis. A weight matrix W (W is an R*R matrix and R is the number of districts considered) indicates the regional interaction. The rth row shows the structure of region r’s interaction with all other regions in the system while Wrs represents for the degree of interaction between region r and s. Zero entries mean that two regions have no interaction, while non-zero entries mean that two regions interact. The weight matrix W is usually standardized by rows that the sum of each row in the matrix equal to unity.
One way of deciding W is to consider the distance between the major cities of the two regions. The first law of geography states that all locations on a map are interrelated and the interaction of closer places is stronger than that of regions far away from each other. Thus, the regional interaction is an inverse relationship with geographic distances.4.4. Dynamic Spatial Shift Share Model
The dynamic spatial shift share model combines the dynamic shift share and spatial shift share models to produce more accurate estimations. With the spatial weight matrix, the employment change is decomposed into four components as SGE, IME, NeCE, and SCE. NeCE is the neighborhood competitive effect and SCE is the state competitive effects which were derived by dividing the CE into two rates .
In equation 3, djk is the shortest road distance between the main cities of LDs of j and k, and is the weighted employment growth rate of sector i in LDj. The NeCE, eijt-1 *(rijt – rijt*) measures the difference between real employment growth and employment growth at the weighted growth rate. When the real growth rate of industry i in the region j is higher than the weighted growth rate of the same industry in the same region, the effect is positive. The SCE eijt-1 *( rijt *– rit) measures the difference between the employment growth at the weighted growth rate and employment growth at the state growth rate for the same industry. When the state growth rate of industry i is smaller than the weighted growth rate of industry i in the region j, the NCE is positives.4.5. Types and Sources of Data
As the study utilizes shift share analysis of employment growth in West Virginia, a panel data set is used for a period of 32 years (1976 to 2007) based on the data availability. The U.S. Bureau of Labor Statistics and U.S. Bureau of Census were the main sources of data. As the interest was to examine the growth of various sectors of local development districts of West Virginia, the data collected on total employment was categorized into ten main sectors (1) farm employment (FRM), (2) mining (MIN), (3) construction (CON), (4) manufacturing (MAN), (5) wholesale and retail trade (W&R), (6) transportation and public utilities (T&U), (7) finance, insurance and real estate (FIR), (8) services (SER): professional, personal and other (9) government (GOV), and (10) agricultural services, forestry fishing and others (AGR). Employment change for West Virginia was considered as the national level in the study. Also, distances between main cities of the local development districts were considered in developing the distance base weight matrix for NeCEs.
5. Empirical Results and Analysis
Table 1 depicts the total employment changes with its components. SGE shows employment growth that would have occurred if the local economy had grown at the same rate as the state economy. Thus, the positive values for all the development districts indicate that all regions could gain in employment if growth occurred at the same rate as the state. The IME measures the amount of local employment sector growth compared to the state level. It can be used in identifying the fast or slow growing sectors in an economy. The positive values for LD1, LD2, LD3, LD4, LD5, LD7 and LD9 mean that those regions are growing faster than the state economy. The negative values of IME in the other districts indicate slower growth compared to the state employment growth. The NeCE shows the difference between the real employment growth and the employment growth at the weighted growth rate. Positive effects of LD3, LD4, LD5, and LD8 indicate that the total growth rates of the districts are higher than the weighted growth rate of the same region. In other words these districts are leading forces in terms of total employment growth. The SCE measures the difference between the employment growth at the weighted growth rate and employment growth at the state growth rate for the same industry. A positive value of SCE means that the local economy has been successful in attracting investment to a particular sector. Moreover, a positive competitive effect indicates the comparative advantage for a region in a particular sector. Thus, the positive values of LD9 and LD11 indicate the potential competitive advantage compared to the other development districts which are negative in SCE. Finally, employment change indicates local development districts of LD3, LD7, LD8, and LD9 had an employment growth while the others had an employment decline within the study period.
Table 2 indicates the state growth effects of each development district with sectors. All sectors of all the development districts are positive in value indicating the potential employment changes if the local sector had grown at the same rate of state economy. In all local development districts service sector indicates the highest contribution to the SGE. In many development districts, the next highest contributing sector is the whole sale and retail trade. Mining, one of the leading sectors in West Virginia seems to be significant only for LD1, LD2, LD3, LD5, and LD6. Contribution of the agricultural sector to SGE is very low in all LDs within 1976 to 2007.
Industrial mix effects of each sector of LDs are shown in Table 3. In all sectors of all LDs, the service sector indicates a fast growth of employment compared to the state level in the last 32 years. Farming, mining, manufacturing, wholesale and retail, transport and utilities, and government service sector indicate negative IME for all LDs, slow growth of employment compared to the state level. Local development districts of LD6, LD7, LD8 and LD9 indicate positive values of IME for construction sector. Except LD10 finance, insurance and real estate sector also indicates a fast growth. The agricultural service sector shows positive values in LD10, and LD11.
Neighborhood competitive effect (NeCE) of each LD is shown in Table 4. LD1 indicates positive values for the sectors of farming, manufacturing, and government sector. The values are positive as the real growth of these sectors is higher than the weighted growth rate of neighbors. Thus, these sectors are the leading sectors for LD1for the period from 1976 and 2007. LD2 has a lead in mining, construction, manufacturing, transport and utilities while LD3 leads in services, farming and mining than its neighbors. Likewise each LD has leading sectors indicating a fast growth within the study period. The negative values of NeCE indicate that real employment of the LD is slower than the weighted average growth of the neighbors.
Table 5 shows the state competitive effect for all sectors. High positive values of wholesale and retail trade, construction, transport and utilities of LD1 indicate the success in attracting investment to these sectors. Negative NCEs indicate that those sectors are losing their shares to other development districts of West Virginia. According to Table 5, mining, wholesale and retail trade, and construction sectors have been the most successful sectors of LD2 in the last 32 years. The positive values of construction, wholesale and retail trade, transports and utilities and agricultural service sectors indicate the economic gaining sectors of LD3. Regarding LD4, SCE values are positive for mining, farming, construction and agricultural services sectors indicating that those sectors have been successful in attracting investments. Positive SCE for farming, mining, construction, and agriculture sector indicate that those sectors have been successful in attracting investments to LD5 for the period from 1976 to 2007. With the exceptions of finance insurance and real estate, service, and government sector, all other sectors in LD6 have been successfully contributing to the economic gains of the district. SCE values of LD7 indicate that only finance insurance and real estate, and service sectors have been losing shares to the other regions in the last 32 years. LD8 results indicate that finance insurance and real estate, service, and agriculture sector have negative values for SCE. Government services, manufacturing, wholesale and retail trade, and mining have been the most attracting sectors for LD9. LD10 has been attracting investments to mining and construction sectors successfully while LD11 has been successful in mining, services, and transport and utilities.
Table 1, Table 2, Table 3, Table 4, and Table 5 show employment changes in local development districts of West Virginia for the period of 1976 to 2007. Results clearly indicate employment growth in some LDs while others indicate a decline. Sector basis analyses show the importance of certain sectors in employment growth of each LD which can help develop better development planning. For instance, in LD1 agricultural services, transport and utilities, services, mining and farming have been contributing the employment growth from 1976 to 2007 (Table 6). Thus, investing in these sectors of LD1 would be more fruitful with the understanding of relevant niche markets. For instance, in the service sector, which includes the services of business, health, engineering, management, professional services, hotels, and personal services, scrutinizing the most effective industry could benefit more economic gains. Moreover, in LD2 agriculture, government, manufacturing, and transport and utilities have been playing major roles in employment growth (Table 6).
Results indicate that mining, one of the major employment sectors, has been gaining in some LDs (LD5, LD7) even though employment decline in the mining sector have been reported for West Virginia within the period of 1976 and 2007 (Table 6). However, the industry currently faces a significant level of regulatory risk related to concerns about climate change and water quality . The manufacturing sector reports employment declines in all LDs in the study period indicating the importance of revitalizing prevailing policies. However, according to the economic outlook for 2011 of West Virginia, there is a potential for employment gains in durable manufacturing sector. But the non-durable manufacturing sectors are losing jobs especially for chemical products . The farming sector in West Virginia has known an employment decline during the past three decades, but LD1, LD3, LD5, LD6 and LD7 indicate an employment growth within the period. Thus, these LDs are potential gainers of investing in farming in the future. The government sector, service sector, and financial insurance and real estate sector seem to have been the most advantageous sectors for many LDs in the last three decades (Table 6).
6. Conclusions and Policy Implications
The dynamic spatial shift share analysis eliminates some of the problems associated with the traditional comparative static approach and provides a more accurate allocation of employment changes among the three shift share effects. The technique is appropriate in measuring the economic effect of various regions in West Virginia. In many LDs service, finance, insurance and real estate, construction, and government sectors seem to be the leading contributors of employment growth in the last 32 years and investment in these sectors would result more employments. The conclusion is supported by the economic outlook of 2011 which forecasts that more employment growth is expected to be generated in the health care; professional and business services; and trade, transportation, and utilities sectors in West Virginia within 2010-2015.
Even though mining and manufacturing are two major industries in the economy of West Virginia, significant employment declines, especially in the manufacturing sector suggest that it is not a significant contributor to employment growth. Interestingly, each LD has a few employment gaining sectors, which could be further invested to accelerate economic growth. Identification of investment priorities in the niches of these potential sectors and implementing a comprehensive development policy could substantially enhance the development of West Virginia. The spatial effects of neighboring regions provide interesting insights of employment changes that make easier in investment planning. The notion of comparative advantage could enhance the efficiency in such a policy.
|||Bartik, T. J. (2004). Economic Development in J. Richard Aronson and Eli Schwartz, eds. Management Policies in Local Government Finance (5th ed.), 355-395.|
|||McNamara K. T. (1991). Employment Growth in Indiana: A State and Regional Analysis. Purdue University Corporative Extension Service, EC 660.|
|||Deming, W.G. (1996). Decade of economic change and population shifts in U.S. regions. Regional Economic Changes. Monthly Labor Review 3.|
|||Knudsen, D. C. (2000). Shift Share Analysis: further examination of models for the description of economic change. Socio-Economic Planning Sciences, 34(3), 177-198. doi:|
|||Melachroinos, K.A. (2002). European Integration and the Spatial Dynamics of Manufacturing-Employment Change. Environment and Planning, 34(11), 2017-36. doi:|
|||U.S. Department of Commerce, Bureau of Census (2008). Statistical Abstracts of the United States. Washington D.C.|
|||State Facts Sheet, West Virginia (2008). Retrieved September 5, 2011 from www.ers.usda.gov/statefacts/WV.htm.|
|||Hammond, G.W. (2011). West Virginia Economic Outlook 2011, The Bureau of Business and Economic, College of Business and Economics, West Virginia University, P.O. Box 6025, Morgantown, WV 26506-6025.|
|||USDA (2008). ERS. States Fact Sheets: West Virginia. Retrieved January 5th from http://www.ers.usda.gov/Briefing/WellBeing/farmhouseincome.htm.|
|||Shi, C and Yang, Y. (2008). Review of Shift Share Analysis and Its Application in Tourism. International Journal of Management Perspectives, ISSN:1307-1629, 1(1), 21-30.|
|||Dunn, E.S.(1960). A statistical and analytical technique for regional analysis. Papers of the Regional Science Association, 6, 97-112. doi:|
|||Ashby, L.D. (1968). The Shift and Share Analysis: A Reply. Southern Economic Journal, 34, 423-25. doi:|
|||Fothergill, S. and Gudgin, G. (1979). In Defense of Shift-Share. Urban Studies, 17, 193-210.|
|||Ledebur, L.C. and Moomaw, R. L. (1983). A Shift-Share Analysis of Regional Labor Productivity in Manufacturing. Growth and Change, 14(1), 2-9. doi:|
|||Andrikopoulos A., Brox, J. A. and Carvalho, E. (1990). Shift-Share Analysis and the Potential for Predicting Regional Growth Patterns: Some Evidence for the Region of Quebec, Canada, Growth and Change, 21, 1-10. doi:|
|||Sirakaya, E., Uysal, M., and Toepper, L. (1999). Measuring tourism performances using shift share analysis: the case of South-Carolina. Journal of Travel Research, 34 (2), 55-62. doi:|
|||Esteban, J. (2000). Regional convergence in Europe and the industry mix: a shift-share analysis. ,), 353-364. doi:|
|||Selting, A. C. and Loveridge, S. (1990). Testing Dynamic Shift share. www.Jrap-journal.org 24, 23-41.|
|||Isard W. (1960). Methods of Regional Analysis: An Introduction to Regional Science, 1960, MIT, Cambridge.|
|||Cliff, A.D. and Ord, J.K. (1969). The Problem of Spatial Autocorrelation. Studies in Regional Science, Edited by A.J. Scott. London: Pion.|
|||Cliff, A.D. and Ord, J.K. (1971). Evaluating the Percentage Points of a Spatial Autocorrelation Coefficient. Geographical Analysis, 3, 51-62. doi:|
|||Cliff, A.D. and Ord, J.K.(1972). Testing for Spatial Autocorrelation Among Regression Residuals. Geographical Analysis, 4, 267-84. doi:|
|||Hewings, Geoffrey J.D. (1976). On the Accuracy of Alternative Models for Stepping-Down Multi-County Employment Projections to Countries. Economic Geography,. 52(3) 206-217. doi:|
|||Nazara, S. and Hewings, G. J.D. (2004). Spatial Structure and Taxonomy of Decomposition in Shift-Share Analysis. Growth and Change, 35(4), 476-490. doi:|
|||Mayor, Matias, and Lopes, A.J. (2008). Spatial Shift-Share Analysis versus Spatial Filtering: An Application to Spanish Employment Data. Empirical Economics, 34, 123-142. doi:|
|||Fernandez, M. M. and Menendez, A. J. L. (2005). Spatial Shift Share Analysis: new development and new findings for the Spanish case. 45th Congress of the European Regional Science Association.|
|||Shi, C., Jie., Z., Yang, Y. and Zhang, Z. (2007). Shift share analysis on international tourism competitiveness: A case of Jiansu Province. Chinese Geographical Science, 17(2), 173-178. doi:|
|||Barff, R. A. and Knight III, P.L. (1988). Dynamic Shift Share Analysis. Growth and Change, 19 (2), 1-10. doi:|
|||Harris, T.R., Gillberg, C.B., Narayan, R., Shonkwiler, J.S. and Lambert, D.K. (1994). A Dynamic Shift Share Analysis in Nevada Economy. Technical Report, UCED.|
|||Markusen, A.R., Noponen, H. and Driessen, K. (1991). International Trade, Productivity, and U.S. Regional Job Growth: A Shift-Share Interpretation. International Regional Science Review, 14(1), 15-39. doi:|
|||Wilson, P. and Ting Su Chern, T.S. (2005). A Dynamic Shift Share Analysis of the Electronics Export Market 1988-2001: Can the NIEs Compete with China? Department of Economics, SCAPE Working Paper Series, Paper No. 2005/07-May2005. Retrieved May 10, 2011, from http/nt2.fas.nus.edu.sg/es/pub/wp-scape/0507.pdf.|
|||Herath, J., Gebremedhin, T. and Maumbe, B. (2011). A dynamic shift share analysis of economic growth in West Virginia, Journal of Rural and Community Development, 6 (2) 155-169.|
|||Quintero, J.P. (2007). Regional Economic Development: An Economic Base Study and Shift-Share Analysis of Hays County. Texas State University-San Marcos, Dept. of Political Science, Public Administration.|
|||Hustedde, R. J., Shaffer, R. and Pulver, G. (2005). Community Economic Analysis: A how to manual. Ames: Iowa State University Printing Services.|
|||Wang. J. (2010). A Dynamic Spatial Shift-Share Analysis of Regional Employment Change in China, SRSA Conference, Now Orland, USA.|