Frequently Asked Questions and Background Information
Direct effects represent the amount of business income earned in the industries in which the expenditures being analyzed were initially made. For example, when analyzing the impact of a museum’s annual expenditures of $1,000,000, the input-output model will show a direct impact of $1,000,000 in the “Museums and historic sites” sector. For spending by non-local visitors, we use visitor counts provided by the organization, along with estimates of per-visitor spending provided by Americans for the Arts, and we divide the non-local visitor spending among several different sectors typically including “Food services and drinking places,” “Hotels and motels,” “Miscellaneous store retailers,” and “Gasoline stations.” Therefore, each of those sectors will typically show some amount of direct effects, which are the estimated amounts of non-local visitor dollars attributed to each category.
Indirect effects result from inter-industry transactions across the supply chain of goods and services that undergirds expenditures in the direct effects categories. As spending occurs in the direct effects categories, other local sectors experience income growth as those sectors respond to the new demands for their goods and services. All indirect effects are the result of spending by firms, not by households, and we utilize the county as the local geographic unit for all indirect estimates since that is the smallest geographic unit for which the necessary data is collected. The input-output model takes account of the propensity for different sectors to purchase their inputs locally, and the share of spending in each sector that goes to households, in determining the magnitude of the indirect effects.
Induced effects are changes in income in various sectors as they respond to new demands from households with increased income as a result of the income that accrues to firms through both direct and indirect. As households spend their income locally, additional income is generated at places like real estate offices, auto dealerships, doctors’ offices, movie theatres, gas stations, and restaurants.
The number of jobs attributable to the combined direct, indirect and induced effects is the sum of the total number of estimated jobs (both full-time and part-time) that are associated with the total impact in each sector of the local economy. The employment effect in each sector will typically be spread across many firms operating in a given sector. The input-output model utilizes data from the ES 202 database, which is compiled from company-level unemployment insurance reports, to estimate the employment patterns for each sector. The figure should not be interpreted as full-time equivalent jobs.
Visitors to a particular cultural organization who live outside the county in which that cultural organization is located. If a single individual makes three separate visits to the organization, we count that as three visitors because each separate visit is likely to entail a certain amount of local spending associated with the visit. The default number of non-local visitors is generally an estimate that is derived in consultation with the cultural organization being analyzed, using address or zip code information obtained from a sample of visitors, or based on anecdotal information. The input figure can be adjusted to account for hypothetical increases or decreases in visits by non-local residents. For the purposes of calculating regional economic impact, we only consider non-local visitors because of what economists call the “substitution effect.” It is assumed that local visitors would most likely still attend other entertainment activities even in the absence of this one particular organization. So, on average, we should assume that their local spending patterns are not significantly influenced by the presence of a particular cultural organization.
The model assumes that the dollar figure entered in the Budget box corresponds to the year entered in this Price Level box. The average per-visitor spending by non-local visitors is automatically adjusted for inflation to the year chosen in the Price Level box. We recommend that users estimate budget expenses and non-local visitors for the current year or most recent year for which such estimates would be reasonable, and fill in the appropriate year for the price level. In other words, if you choose 2006 as the Price Level, then it is assumed that you entered 2006 budget and non-local visitor figures, and the impacts given are in 2006 dollars.
More FAQ for the Regional Economic Impact Model
Who developed this tool?
The tool was developed by researchers at the Center for
Creative Community Development (C3D). The primary researcher was Dr.
Stephen Sheppard, Professor of Economics at
How confident can I be of the economic impact estimates?
The economic impact estimates are based on standard input/output analysis. This type of model has been in use at least since the publication in 1960 of Walter Isard’s important book Methods of Regional Analysis: an Introduction to Regional Science (M.I.T. Press).
What would I use this tool for?
The regional economic impact calculator can be used to obtain rough estimates of the economic impacts resulting from a given amount of annual spending by a nonprofit cultural organization and from the estimated local spending of non-local visitors to that cultural organization. The default values are intended to provide an approximation of the estimated impact in 2008 based on information obtained from the organization. However, you can estimate impacts that would result from increasing or decreasing the annual operating budget of the organization, or from increasing or decreasing the number of non-local visitors to the organization. Many arts administrators value this kind of economic impact information for the purposes of talking with donors, policy makers and local constituents about the importance of their efforts. The tool can also be useful in strategic planning to evaluate alternative scenarios, looking at the relative change in estimated impact associated with various programs or outcomes. This tool is not intended to be used to estimate impacts of cultural organizations in other locations not specified here because each organization’s model reflects the nature of the local economy in the specific county where the case study organization is located.
What do you mean by a model?
Models are sets of mathematical formulas whose values are based on statistical analysis of actual observations. In this case the formulas are designed to represent the working of the local economy. The economic impact estimates provided here are the result of a predictive model that estimates the amount of aggregate local income and employment that is attributable to expenditures by a particular cultural organization and its non-local visitors. The model used to generate this result is a traditional regional input/output model.
What is the region covered by the economic impact estimates?
For both local employment and aggregate local impact estimates, the region used in this model is the County. The total impact given is the estimated amount of aggregate local income and the estimated number of local jobs in the County that can be attributed to expenditures by a particular cultural organization and its non-local visitors (visitors living outside the county).
How do you estimate the amount that is spent by non-local visitors?
In 2005, Americans for the Arts (link to www.artsusa.org) engaged in extensive nationwide surveying of visitors to cultural organizations of all types and sizes. A total of 94,478 visitors at cultural organizations completed surveys about their party’s expenditures directly related to making that day’s visit. As reported by Americans for the Arts, “the randomly selected respondents provided itemized expenditure data on attendance-related activities such as meals, souvenirs, transportation, and lodging. Data was collected throughout the year (to guard against seasonal spikes or drop-offs in attendance) as well as at a broad range of events (a night at the opera will typically yield more spending than a Saturday children’s theater production, for example).” Except in cases where we feel there is better data available, we typically rely on Americans for the Arts’ national average for our analysis, which is $40.58 per non-local visitor, including: transportation (gasoline) $5.90; souvenirs and other retail purchases $11.11; lodging $8.02; and restaurant meals $15.55. Our tool adjusts these figures for inflation depending on the price level year selected. It should be noted that although $8.02 is much lower than the cost of a room at any hotel, that figure is an average that accounts for many respondents with no reported lodging expenses (day-trip visitors) and a smaller number of respondents with lodging expenses much higher than $8.02. Also, remember that respondents were asked to only report expenses directly related to their visit that day: if the respondent did not attribute her lodging solely to her visit to that particular cultural organization (because she had multiple reasons for visiting the area), then she may have counted little or none of her lodging expense on her survey response. That helps make these estimates conservative and defensible from an economic impact perspective. The Americans for the Arts report is titled “Arts and Economic Prosperity III: The Economic Impact of Nonprofit Arts and Culture Organizations and Their Audiences,” which may be downloaded at www.artsusa.org/information_services/research/services/economic_impact/default.asp.
Would the impact be significantly different for the individual town or city in which the cultural organization is located, or for the entire state, as opposed to the county?
Unfortunately, no government or private entity collects the data that would be necessary to do an input-output economic impact analysis restricted to impacts for a single town or city? It is logical to assume that much of the estimated local impact at the county level does occur in the town or city in which the organization is located, but the share could vary widely depending on the structure of the local economy.
Our input-output model can estimate impacts at the state level, but we have not constructed our interactive tool to perform state-level analysis because that is not usually the preferred geographic range for economic impact reports. The state level analysis would take into account the flow of goods and services within the state as a whole. Total economic impacts, when viewed at the state level, can be either higher or lower than those experienced at the county level. Two things happen as we move from a county-level analysis to a state-level analysis. First, because the state economy is so much larger than a single county’s economy, we expect expenditures of both the cultural organization itself and its visitors to circulate longer within the state economy before ‘leaking’ to other states. This will tend to make the state-level impact larger. On the other hand, you have to take account of a larger “substitution effect.” If the organization’s visitors include many people who live outside of the county in which the organization is located but in the same state in which the organization is located, then the number of visitors who will be termed “non-local” at the state level (i.e., visitors who live not only outside the county but outside the state) would be smaller than the number of non-local visitors when viewed from the county perspective. This shrinkage in what we consider non-local visitation will tend to reduce the state-level impact of visitors relative to the visitor impact at the county level.
These two factors may roughly cancel each other out, or one may weigh more strongly than the other, depending on the structure of the local economy and the nature of the organization’s visitor demographics.
Is the employment estimate based on “full-time equivalent” jobs?
No, our estimate does not translate part-time jobs into full-time equivalent jobs. The input-output model utilizes data from the ES 202 database, which is compiled from comprehensive company-level unemployment insurance reports in every state, to estimate the employment patterns for each sector. In each county, the total amount of output in a given sector is associated with a total number of jobs (whether full-time or part-time) in that sector as demonstrated by ES202 data, and that ratio determines the estimated total number of jobs attributable to each dollar of output in that sector. Our “jobs” estimate is the sum of the estimated employment impact in each sector.
What underlying data is the model based on?
The input-output model utilizes data from a variety of sources (including the U.S. Bureau of Economic Analysis, the U.S. Bureau of Labor, and the U.S. Census Bureau) to characterize the flow of goods and services among sectors of the economy and the employment and consumption patterns of different sectors of the local economy. The sectors are identified by NAICS (North American Industry Classification System) codes. Much of the data is collected at the county level through a survey process that examines the spending patterns of representative firms in every sector of the economy in every county throughout the country. The data collected includes estimates of the purchasing patterns of each sector of the county economy, identifying how much of every dollar spent in one particular sector is received as income in every other sector of the county economy, and how much of every dollar “leaks” outside the county economy or is considered “final consumption.” The input-output economic model divides the economy up into over 400 sectors ranging from “Abrasive Products” to “Wood window and door manufacturing”. Not all of these sectors are present in every local area. The model also draws heavily on data from the federal ES202 database of unemployment insurance filings and the “Regional Economic Information System” of the U.S. Bureau of Economic Analysis.
Does this tool involve a “multiplier”?
The model described above can be understood to produce a
multiplier that is distinct for each of the cultural organizations evaluated,
and depends upon the nature of the local (county) economy. In other words, a museum in
It should also be noted that our model assumes the impact of a particular cultural organization to result from direct spending in more than 1 sector. We generally assume that all of the spending by a museum can be classified as occurring in the Museums and Historical Sites sector, but occasionally we may estimate that it would be more appropriate to divide a particular cultural organization’s spending among two or more sectors such as “Museums and Historical Sites,” “Performing Arts Companies,” and “Civic and Social Organizations.” Likewise, we always estimate the spending by visitors to a cultural organization to be divided among several different sectors typically including “Food Services and Drinking Places,” “Hotels and Motels,” “Miscellaneous Store Retailers,” and “Gasoline Stations.” The attribution of visitor spending to different sectors may vary from case to case based on the available data about visitor spending habits. By estimating direct spending in multiple sectors, we end up with several different “multipliers” at work, and the total economic impact consists of the combined effect of those different multipliers.
Should I interpret these results to be an exact measure of the economic impact of this organization?
When we hear the phrase “your actual mileage may vary,” most of us expect that this means “your actual mileage may be less or, if you are really lucky, more.” Input/output economic impact estimates should be interpreted similarly – that the impact on income and employment will be approximately this value on average – sometimes less, sometimes more. We have endeavored to make the economic models as accurate as possible, but you should remember that these results are based on average experiences of cultural organizations in a particular sector within a particular county. Individual cultural organizations may generate larger or smaller impacts, depending on the extent to which they purchase their inputs locally, or have non-local visitors whose spending patterns deviate significantly from the averages used.
We urge you to carefully read the background information and advice contained throughout this FAQ so that you can better understand how the results are obtained and how you can correctly interpret and utilize those results. Your economic impact claims will be persuasive if you can articulate why they make sense. The estimates provided by this calculator can be an effective starting point for a case study and more detailed analysis.
Can you give me context for this estimate of economic impact? What is the average economic impact multiplier for the organizations you’ve studied, and for firms in general? How much does it vary?
The economic impact is dependent on the size (expenditures) of the organization, the type of organization and its location, and the size of the non-local visitor audience. The impact varies between types of organizations because different organizations purchase different types of goods and services, and these affect the local economy in different ways. The impact varies between locations because different locations have more or less developed sectors for providing the goods and services needed by an organization, and hence a different local impact is generated.
To illustrate, consider an example. Suppose identical performing arts companies are set up in two different communities. In one of the communities, there is a supplier of special lighting equipment, and in the other there is none. When a performance takes place in the community with the lighting supplier, the performing arts center purchases lighting equipment locally. These purchases pay a local person, who then spends some of the money locally, contributing to the local economic impact. In the other community, lighting equipment is ordered from outside the community. These purchases pay someone outside the community, who spends the income there. It contributes to the broader, national economy, but does not contribute to the local economic impact.
This can generate a wide range of impacts. In the cultural organizations we’ve studied, the total impact has generally ranged between 2.5 times the budget of the organization and 5 times the budget of the organization. This includes the impact of non-local visitors, which have a significant effect on the impact ratio (organizations that attract more non-local visitors per budget dollar, of course, will tend to have a higher ratio of total impact to organizational budget). In the cultural organizations we’ve studied, the ratio of total impact to organizational budget excluding non-local visitors has generally ranged between 1.6 times the budget of the organization and 2.4 times the budget of the organization.
What websites do you recommend for further reading and understanding of the economic and social impacts of cultural organizations?
Wikipedia provides a brief overview of the input/output model utilized in our regional economic impact analyses: http://en.wikipedia.org/wiki/Input-output_model.
Americans for the Arts has conducted extensive research on the spending patterns of visitors to cultural organizations and the economic impact of those organizations. We recommend visiting http://www.artsusa.org/information_services/research/services/economic_impact/default.asp for more information on their work.
New England Foundation for the Arts also has a long history in the “creative economy” field, producing research and advocacy materials available at http://www.nefa.org/projinit/thecreativeecon.html.
Partners for Livable Communities has been a collaborator of ours for the past four years through the “Shifting Sands” program funded by the Ford Foundation. Information on the Shifting Sands grantees and on creative community development strategies in general may be found at http://www.cultureshapescommunity.org/.
The Social Impact of the Arts Project (http://www.sp2.upenn.edu/SIAP/) at
The Pew Charitable Trust has overseen extensive state-wide
data collection and analysis in the nonprofit cultural sectors in
How can I create a screen-shot to insert into a word document or slide presentation?
For the best screen-shots, first press F11 to make the web page full-screen (and you can press F11 again when you’re done with your screen shot to bring back the menu bars). When you have the screen image you want, just press the “Print Screen” key to save the screen-shot to your Windows clipboard. Now choose Start, Programs, Accessories, Paint, and when that application opens choose Edit, Paste to place the image into the workspace. You can now edit the image using any of the tools available, then choose File, Save As to save it as a file in an appropriate location on your hard drive.
Draft Glossary of Key Terms for the Visitors/Friends/Members Origin Map
Census Layer: The interactive map can display a single census variable at a time, while also displaying the locations of visitors, friends, members or other organizational constituents. The census layer dropdown menu is used to select from a list of more than 20 data fields from the 2000 census, with color-coded results presented for every census block-group within the census range (as defined above).
Visitor/Friend/Member Layer: If we have mapped multiple address lists for a single organization, then individual lists may be turned on or off by selecting from the visitor layer menu.
Geocode: To apply geographic coordinates to a street address. This feature allows for customized mapping of address lists provided by the user (see FAQ for more information).
DRAFT FAQ for the Visitors/Friends/Members Origin Map
How was the visitor/friend/member data collected?
We have mapped the visitor, member or other address data as provided to us by each organization. Some organizations use box office software to collect the addresses of the majority of visitors. Other organizations have collected information in comment books and then digitized the addresses provided. In some cases, organizations have conducted surveys of visitors to collect sample data about their visitor locations and have asked us to map the results. The organizations never obtain 100% coverage of their visitor addresses, but the use of box office software is generally associated with higher representation. Member data, on the other hand, generally provides 100% representation of the actual member base since virtually every organization maintains mailing address data for its members.
What would I use this mapping tool for?
The visitor map with layered census data can serve many purposes. On the most basic level, a visitor map is useful in presenting the geographic range of an organization’s visitors with eye-catching imagery that can be inserted into reports or proposals.
Many arts administrators or policy makers would also like to be able to better illustrate the composition of the audience for a particular arts organization because they define that organization’s social impact, in part, as delivering important cultural and educational experiences to members of diverse and/or disadvantaged communities. The visitor maps allow you to display the economic and ethnic characteristics of the communities from which an organization’s audience is drawn, and can sometimes reveal pockets of visitor engagement that might otherwise go unnoticed. Perhaps an organization has made concerted efforts to attract visitors from low-income neighborhoods. The “% Poor” layer, displayed in conjunction with an appropriate visitor address list, would help illustrate whether that effort had met with some success. Over time, an organization might use the mapping tool to compare address lists before and after a particular program had been undertaken to show how the constituency had evolved in relation to new programming.
Even a simple visitor map may have more power of persuasion than a chart or graph based on the same data, or at least can complement other forms of data presentation.
What does the “% of Addresses by
To augment the visual storytelling of the maps, this feature reports the frequency with which visitor addresses plotted on the map fall within the percentage range categories provided in the key. This provides important context for interpreting the information on the map. For example, when viewing the “% Poor” layer, you might note that many of the census block-groups in the census range show poverty levels in the highest range on the key, but when you view the “% of Addresses by Census Range” window you see confirmation that only a small percentage of this particular sample of visitors to the organization live in those census block-groups with high poverty. This would suggest that the organization is not engaging substantial numbers of visitors from low-income neighborhoods as a share of its overall audience.
What does the “Geocode” feature do?
To provide customized mapping, this feature allows users to input a set of addresses and then plot those addresses on the map as a new data layer, with color coded “flags” to show the locations of individual addresses.
How does the local data shown on the map compare to state or national averages?
For an overview of the census results for a particular county or state, visit http://factfinder.census.gov. Comparisons to broader averages do indeed provide helpful context for analyzing the map.
How can I create a screen-shot to insert into a word document or slide presentation?
For the best screen-shots, first press F11 to make the web page full screen (and you can press F11 again when you’re done to bring back the menu bars). When you have the screen image you want, just press the “Print Screen” key to save the screen-shot to your Windows clipboard. Now choose Start, Programs, Accessories, Paint, and when that application opens choose Edit, Paste to place the image into the workspace. You can now edit the image using any of the tools available, then choose File, Save As to save it as a file in an appropriate location on your hard drive.
DRAFT FAQ for the Impact on Property Values Tool
Because having cultural activities and the arts available in a community makes the community a more pleasant place to live and enhances local educational opportunities. This in turn increases the demand for housing in a community. Increasing the demand for housing in a community tends to increase property values. Good schools, lower crime rates, and environmental quality all have been shown to have this effect. We use standard economic modeling techniques to measure the impact that cultural organizations can have in making communities more attractive.
The property value impact is shown in the form of estimates of the percentage change in market value of houses in the vicinity of a particular cultural organization after some key event has occurred (such as a major expansion to the organization). Our model estimates the organization’s impact on property values after screening out changes in value associated with general market conditions. Economic theory asserts that area-wide changes in property values are tied to changes in the quality of life in the surrounding neighborhood (in addition to changes in economic conditions in the area). When our modeling identifies neighborhoods experiencing significant changes in property values that are mathematically attributable to the presence or expansion of a particular cultural organization, we interpret that to reflect a change in the quality of life in that community due to the existence of the cultural organization.
The property value impacts estimated by our models are best thought of as the change in the value of housing after the market adjusts to the opening or expansion of a particular cultural organization. Think of a town in which house prices were changing at exactly the average rate for the region. An expansion by an important cultural organization improves the quality of life in the community and increases the demand for housing and hence house prices. How long this takes depends on how active the local house market is, how quickly people become aware of the improvements, and how confident they are that the change is permanent. A time period of 6 months to a year is probably the minimum, while 3 years is probably the maximum. After this period of adjustment, local property values would have adjusted to a new level, and (assuming no further changes) local house prices would again tend to follow the regional trend.
Estimates of the impacts of community characteristics vary widely. Some studies of the impact of school quality suggest that moving an identical home from the worst school to the best could change the value by as much as 30%. Studies of the impact of open space preservation suggest that open space within walking distance of a house can increase its value by about 10%. The analysis we’ve done of the impacts of cultural organizations in various locations suggests that, in certain markets, a major cultural investment (valued in the millions of dollars) can sometimes increase the value of nearby properties by more than 15%, with decreasing impact as you move farther from the organization.
The impacts are estimated from ‘models’ that are mathematical formulas whose values are based on statistical analysis of the actual sales prices of thousands of houses in the local community, relating the sales price to the characteristics of the house, the proximity of the house to the cultural organization, and the timing of the sale compared to the timing of the opening or expansion of the cultural organization, while incorporating numerous other factors relevant to the local housing market.
Yes. The models we use account for many factors such as size, condition, style and age of house, location relative to town center, lot size, and factors that vary between towns but tend to be stable over time (such as school quality, traffic patterns and environmental quality). In addition to all these factors, we examine the impact of proximity to a particular cultural organization that has undergone dramatic change over a certain period of time.
This depends on the application and interpretation. The models we use are generally successful in explaining about 70-75% of the variation in house prices within the communities studied. This means that factors for which we have limited (or no) information such as curb appeal, landscaping or interior layout account for the remaining 25-30% of value. For an individual house, the models we use do a good job, most of the time, in predicting the market value (sales price) of a house. There are cases, however, where the model does not predict a particular house value correctly, particularly when the house has unique characteristics. The best interpretation is to think of our model’s estimated impact as an average impact that would apply across a range of houses that share certain characteristics in common (most importantly, proximity to the cultural organization). Even within broad bands of estimated impact, there will be variations in the actual impact that can’t be observed in our models, so please always remember that these are estimates.
Our models focus on changes in house prices relative to the regional trend in house values during a certain time period. Thus the models only count a property value increase as having occurred if the price went up by more than the average increase in the region. In that way, the changes attributable to a cultural organization are not dependent on broad market trends, especially the boom and bust of the last 5 years; in most cases, our analysis is focused on impacts that were observed well before the most recent boom and bust.
Our models do not consider non-residential property values. We focus on the impact on residential property values for several reasons: First, the changes in residential property values provide an important barometer of the impact that cultural organizations can have on the quality of life in a community. Second, changes in property values can pose particular policy challenges in communities where many or most residents are renters. Ensuring continued access to culturally vibrant communities for all households is an important goal for communities. Finally, there are fewer observed sales of commercial property, so that statistical analysis of the impacts of a cultural organization on commercial property values involves greater uncertainty.
DRAFT FAQ for the Social Network Map Tool
Social network data can be collected in a variety of ways. The simplest method is for an organization to create a list of its key partners (with street address), and characterize the nature, strength, and/or longevity of the partnership to add richness to the map. An alternative method is for an organization to survey its staff, board, and volunteers about their other organizational affiliations and map the social network on the basis of these individual and organizational linkages formed by persons affiliated with multiple organizations. A third approach is to reach out directly to organizations of every kind in a certain geographic area and ask all of them to share lists of employees, board members and/or volunteers. That can be very time-consuming, and even with significant effort it can be difficult to obtain a nearly complete set of employee/board/volunteer lists in a specific geographic area. The effort can be limited to registered nonprofit organizations to simplify the task, but even that can still be daunting.
Any social network dataset will only be as comprehensive as the organization is able to make it with available time and other resources, so it is not always fair to compare the social network maps of different organizations that may have been compiled through widely varying methods or with greater or lesser effort to be comprehensive.
A social network map allows an evaluation of the extent to which a community is interconnected and information can flow between many organizations, and the extent to which an individual organization might play a key role in the flow of information and resources between organizations. An organization’s social network can be an indicator of whether that organization is a ‘coalition builder’ (with an array of links), for example, or ‘isolationist,’ (with few links to other organizations in the community).
The results of social network analysis can be a resource for the cultural organization in communicating its involvement in the community. For instance, with a social network map, an arts organization could demonstrate its collaborative potential in grant proposals.
Additionally, the results of social network analysis could be part of strategic planning within the organization. With a map of its social network, a cultural organization might better assess the community resources available to it. It could examine the types of organizations to which it has access, and that have access to it, and potentially reach out to new types of organizations.
Network analysis has a real benefit when done on the entire community. This allows the cultural arts organization to bring something of community value ‘to the table,’ and can help initiate a community-wide discussion about identity and goals. ‘Before’ and ‘after’ network maps can be used as an evaluative tool for program initiatives with a goal of building new coalitions or of making the community more cohesive.