Section 5: Measuring the regional economic linkages of the child care sector

from Measuring the Regional Importance of Early Care and Education: The Cornell Methodology Guide, by Rosaria Ribeiro and Mildred Warner, Ph.D.

Child care’s importance as an economic sector stems not only from its direct employment and output in the regional economy, but also from its linkages to other industries. From a regional economic standpoint, the most important measure of a sector’s economic importance is the size of employment and output. However, the regional economic impact of a sector involves more than simply its size; each industry also has a linkage effect in the broader regional economy. The regional economy is composed of many industries that buy and sell from each other. These inter-industry purchases can be measured to show the relative strength of inter-industry linkages for each industry in the regional economy. Child care businesses and employees purchase goods and services that stimulate economic activity in other industries. Economic impact analysis makes it possible to estimate the dollar value of these linkages.

This section provides an overview of regional economic modeling, known as input-output analysis, and how it can be used to better understand the economic contribution of the child care sector. A simple model of the regional economy is provided below. As money circulates between industries in the regional economy, it stimulates economic activity. These activities can be considered “ripples” in the regional economy pond. Money leaks out of the regional economy through savings or purchases made outside the region. The ripple effect is larger when the leakages are smaller.

INPUT-OUTPUT ANALYSIS

Input-output models are based on the assumption that export demand (or the ability of industries to sell to the external economy) is the engine that generates activity in the regional economy. Changes in final demand¹ (direct effects) infuse local industries with new funds, which increase output and employment. There are two types of linkage effects that input-output analysis measures:

• Indirect effects count the multiple rounds of inter-industry purchases spurred by child care industry spending. Child care businesses purchase food and supplies from other industries, in turn stimulating output in those industries.
• Induced effects capture the impact of household spending. Employees spend their wages in the larger economy and these expenditures generate demand in other industry sectors (housing, groceries, etc.).

The direct, indirect and induced effects of the child care sector are shown in Figure 6.

Multipliers, or “linkage effects,” can be calculated for output, employment, labor income and value-added; however output and employment multipliers are the most common and most easily understood.

• An output multiplier for the child care industry estimates the total sales that would be generated in the entire economy by each dollar of increased direct spending for child care services.
• The employment multiplier is an estimate of the gross number of jobs that would be created throughout the regional economy from an increase in demand for child care services large enough to stimulate the addition of one new job in the child care industry.

Generally, the Larger the Economy, the Larger the Linkage Effect

Multipliers express the degree of interdependence between sectors in a region’s economy and therefore vary considerably across regions and sectors. Typically, larger economies have larger multipliers because they are more self-sufficient than smaller economies, therefore the leakages are smaller. In Kansas, the Kansas City Metro region’s multipliers were almost as large as the state as a whole (see Figure 7). Economic activity within the metro region is highly interconnected – leading to large linkage effects. Agriculture is an industry with strong vertical integration in the rest of the state, but much of the linkage for other sectors occurs outside the state.

Multipliers measure linkage – and size is only a rough proxy for economic structure. It is also possible for smaller, isolated economies to have large multipliers because there is less leakage to the outside world. For example, in the New York State study, Long Island had larger multipliers than the five boroughs of New York City.

IMPLAN Modeling Program

Regional economic modeling known as input-output analysis can be used to measure the linkage effect of any industry. The IMPLAN modeling software is the most commonly used program for the regional economic analysis of the child care sector. IMPLAN allows the user to build economic models to estimate the effect of economic changes in a state or region. The program includes data for 528 business and industry sectors (4 digit SIC in manufacturing and 2-3 digit for other sectors) including child care. The data used by IMPLAN are primarily from federal sources including the Input-Output Accounts from the Bureau of Economic Analysis (BEA), the Covered Employment and Wages Program (ES202) from the Bureau of Labor Statistics (BLS), and others (IMPLAN 2002).

COMPARING THE CHILD CARE SECTOR WITH OTHER ECONOMIC SECTORS

Multipliers help policy makers understand the different linkage effects associated with different industries. For example, policy makers might want to know if allocating funds to the child care industry is likely to produce more or less regional economic linkage than investments in other infrastructure sectors such as job-training programs, education, water and sewer or transportation.

One of the uses of multipliers is for comparison between industries. In the Kansas study, child care output multipliers were compared with those of other important Kansas industries, as shown in Figure 8.² Child care multipliers were higher than retail or lodging – economic sectors which typically get economic development attention. The output linkages for child care reflect the fact that most of the child care industry’s purchases are local and these inputs are likely to be produced locally. The retail industry, by contrast, purchases many of its inputs from outside the local economy – which creates a leakage. The meat packing industry purchases many of its inputs locally – a reflection of the vertical integration of Kansas’ agriculture industry and its importance as an economic driver for the state. Across states, comparisons between child care multipliers and those for agriculture or manufacturing are more variable – reflecting the different composition of regional economies due to specialization.

In Kansas and in other states, we have found child care has similar multipliers to other infrastructure sectors: education, job training, transportation. Figure 9 shows child care employment multipliers are slightly higher than in other infrastructure sectors, except water and sewer (this reflects the capital intensive nature of water and sewer relative to the other infrastructure sectors). Job training, education and physical infrastructure are typically viewed as worthy of public tax-based expenditure, both for the intrinsic value of their products and for their economic development impact, while child care is rarely considered in economic development terms. However, output and employment multipliers show similar strength of inter-industry linkage for the child care sector.

Milwaukee used the output and employment multipliers to show which industries would be affected by a change in child care demand. Calculating the indirect labor effects on specific sectors in the regional economy can be very helpful for economic development planning purposes. This technique can be used to project changes in labor demand in other sectors if there is an increase in demand in one sector (Shideler and Fikkert, 2001).

USING INPUT-OUTPUT MODELING TO ASSESS CHANGES IN THE REGIONAL ECONOMY

Input-output modeling is used to measure the impact of a change in demand. Changes in demand act as a “shock” to the economy and the input-output analysis tells us the full impact of that shock (on both the sector in question and the other sectors to which it is linked). Multipliers should be used only on net changes to the economy that increase or decrease final demand. Some studies multiplied the gross receipts and employment of child care by its output and employment multipliers to estimate the total economic impact of the sector. This method is not appropriate because it treats gross receipts as equivalent to final demand, which over counts the impact. Multipliers should be used as a tool to help calculate the economic impact of changes in demand, or viewed on their own as a measure of the strength of inter-industry linkages. A loss/increase in government funding would be an example of a change in final demand.

Use Type I multipliers for changes in internal (regional or local) demand.  
Input-output analysis normally assumes that demand originates outside the local economy. However, households are the primary purchasers of child care, and demand is usually local. Type I multipliers treat households as external (exogenous) to the economy being modeled. Type I multipliers include only the direct effects of the child care sector and the indirect effects of inter-industry purchases. (see Figure 10). Child care businesses typically purchase more locally than other industries and these business linkages are captured in the Type I multipliers. Type I multipliers do not include the induced effects of household spending. To increase their spending on child care, households would have to reduce their spending on some other sector, and thus, we do not attempt to calculate induced effects on changes in internal demand.

In a state level regional model, state funds form part of the demand from inside the region. In 2003 New York State was facing a serious fiscal crisis and one of the scenarios in the 2003 state budget debates was to eliminate the state funded pre-kindergarten program. A cut of this magnitude would create a shock to the early care and education system. The total effects would extend beyond the direct cuts in employment and income to the child care sector. Reduced spending by the child care sector would have a linkage effect on industries that supply food, furniture and other goods to the child care sector. The direct effect of this shock was estimated to be $204 million in state funds and 6,000 workers. To determine the indirect effects, a Type 1 multiplier was used to show an additional loss of $106 million ($204*.52) for a total loss of $310 million to the state economy. Similarly, the indirect effects of the employment loss were 1,620 jobs (6,000*.27) for a total job loss of 7,620 jobs. Fortunately, the State was able to find funds to keep the program going.

Use Type II multipliers for changes in external demand.
Type II multipliers include the direct, indirect, and induced effects and are used to calculate the impact of a change in external demand for child care (see Figure 11). The primary source of external demand for child care is federal investment in a state’s child care sector. We can presume that federal funding is an external investment that supplements household demand for child care. Because this shock can be treated as net new demand to the regional economy, we can include both the indirect effect of industry purchases and the induced effect of household and worker expenditures in calculating the total linkage effects.

For example, Head Start funding is primarily federal. These federal funds supplement parental demand for child care. In New York State in 2002, $420 million in Federal funds were spent on Head Start programs. Due to the indirect and induced linkage effects, these federal funds have a total regional economic effect in the State of $857 million ($420*2.04 Type II multiplier).

There is some debate among economists about whether it is appropriate to use Type I or Type II multipliers when analyzing shocks to the child care sector.³ We used Type II multipliers on external demand, e.g. federal aid in the New York State model (the Head Start example above). In the pre-kindergarten example above, a Type I multiplier (which assumes no change in household expenditures) was used for the state funds. However, one could assume that if state funding for pre-kindergarten had ceased, this would have stimulated some changes in household expenditure. Study teams should carefully consider which approach is most appropriate for their context, recognizing no optimal solution is available given the importance of household demand, state and federal funding to the child care sector. Some study teams have decided to put most of the detail on their linkage analysis in an appendix due to unresolved debate about how to handle the child care sector in an input-output framework.

METHODOLOGICAL CHALLENGES OF USING INPUT-OUTPUT MODELS

Input-output models are based on the export base theory of economic growth, which argues that external demand – exports – is what drives economic activity. While exports are certainly important, consumer demand now accounts for 63% of US final demand. Exports, by contrast, only comprise 9% of final demand nationally.⁴ Thus, economic models that give primacy to export demand may be missing important internal sources of regional economic activity.

The demand for child care is primarily local – from households. Federal funding for child care has increased dramatically under welfare reform and these federal funds supplement parental demand and serve as a source of external demand in a state’s economy. Thus, using input-output models to analyze changes in federal funding is similar to using them for export-based demand.

The role of household consumption is still important in its own right. Some economists argue we should not count household effects (the induced effects) because households would have spent their money on something else anyway. However, others recognize the importance of local consumer demand for economic development. Local governments are keenly aware of the importance of capturing leakages to strengthen the local economy. This is why the competition for regional retail establishments is so keen. Household demand is an important part of the linkage effect of child care.

Input-output models were originally built to study the impacts of export-based sectors like agriculture, forestry and manufacturing. However with the growth in service sector employment (which now comprises 80 percent of all employment nationally), more attention has been focused on the economic impact of services. Business, financial and information services are now recognized as important sources of economic growth (Drennan, 2002). Personal and consumer services, while not economic drivers, are also leading sources of employment growth in regional economies. Child care, due to its labor intensive nature, is a significant employer and an important service sector which supports parents who enter the labor force.

There is a limit to how well input-output models can measure the impact of a service sector such as child care. Input-output multipliers measure only backward linkages, i.e. the linkage between one industry and its suppliers. The child care sector demands primarily labor, supplies and food for its direct use. The most important contribution of child care is its consumption/investment effect in educating young children. An additional value is the role child care plays in supporting parents who work. Child care’s capacity to support working parents can be termed a “forward linkage.” Forward linkages are not measured in input-output model multipliers.⁵ The Cornell project is working on a procedure to use input-output modeling tools to measure the combined impact of backward and forward linkages. This alternative approach will better capture the regional economic importance of child care. In the meantime, using standard input-output analysis to look at the child care sector is justified, especially if care is taken to distinguish local from external demand.

¹ A sector’s outputs are demanded both inside and outside the regional economy. Final demand in an input-output framework is that portion of demand that is not used in the production of other outputs inside the regional economy (intermediate demand). Final demand includes consumption, investment, government and exports.
² An industry’s economic impact on an area is not only a function of its multiplier, but how much money it is attracting to the economy from outside the region (external demand). Therefore, while it may be appropriate to compare industry multipliers within a broad sector like services, or similar infrastructure sectors, it is less appropriate to compare multipliers between services and manufacturing or agriculture.
³ Type I multipliers assume households do not change expenditures with changes in income while Type II estimates assume households change expenditures linearly with income.
⁴ Economic Report of the President, 2002.
⁵ Some studies have used the input-output model to run multipliers on parent wages and have attributed all of this impact to child care (M.Cubed 2002). We do not recommend this. Parents earnings are not attributable to the child care industry, but to the industries where parents actually work. The support child care provides to parent workers is a forward linkage that is not counted in an input-output multiplier. The Cornell research team is working on a new approach to input-output modeling that accounts for the forward linkages (sales) of the child care sector.