A GUIDE TO DEVELOPING AND USING FAMILY AND
CHILDREN’S BUDGETS
Prepared for
THE FINANCE PROJECT
ABOUT THE AUTHORS
Mark Friedman served for 19 years in the Maryland Department of Human Resources, including six years as the Department's chief financial officer. Aftre four years with the Center for the Study of Social Policy, he established, and now directs, the Fiscal Policy Studies Institute in Baltimore, Maryland. Mark is a member of The Finance Project's Working Group on Results-based Planning, Budgeting, Management, and Accountability Systems.
Anna Danagger is Senior Research
Associate at The Finance Project.
I. INTRODUCTION
A family and children’s budget is a document that summarizes spending for children and their families for a nation, state, county, city or community. This paper is about the development and use of family and children's budgets.
Why create a family and children’s budget? In simple terms, family and children’s budgets can help us get better results for children and families. If we are serious about the well-being of children, then we will be businesslike in our work to improve the conditions of children and their families. We will begin to think about the kinds of investments necessary to produce better results for children and families. And we will begin to build and use the decision-making tools necessary to do this well. Most basic among those tools is a picture of how resources are now being used: How much is spent, by whom, for what? No business would make investment decisions without such a picture. No community trying to do a better job of helping children to develop and thrive should do less.
Making sense of spending for children and families is no easy task. Spending for children and families is spread across different levels of government (e.g., federal, state, county, city, town, school district), across many agencies within each level, and across public and private sectors. It involves dozens of funding sources, paying for hundreds of different programs. And while this system does a good job of meeting the needs of some children, it misses others badly. This fragmented system of funding reflects a more profound fragmentation of services, based on categories of children, categories of service, and a division of responsibility between funders accumulated from years of political deals, more than any sensible way to pay for and provide service.
A family and children's budget is a policy tool that can help unravel this complex system, understand how services are now provided and funded, and make better decisions about how to use our limited resources to advance the well-being of children and families. A family and children’s budget can help answer seemingly simple questions like: How much is spent, for what service, by what agencies? How much are costs increasing or decreasing? How are spending priorities changing over time? And also more complex questions like: Are children receiving their fair share of revenue growth? In times of cuts, are they protected more or less than other parts of the budget? How does our spending for children’s services compare to other similar jurisdictions? Are we using our resources efficiently? What investments will produce the greatest future benefits for child and family well-being and reduced cost of remediation?
In total, we are spending a lot of money on children and families. And a large percentage of that spending is for remediating problems after they occur, not in investing in the healthy development of children and families necessary to prevent or reduce these problems. This means, in simple terms, that we are almost certainly paying more for remediation than we would if we approached the well-being of children as a matter of investment. There is a growing consensus, if not yet a fully conclusive body of evidence, that substantive investments in child development, family support, and prevention is not only good social policy but good fiscal policy as well. Family and children’s budgets can help us understand our choices and act on our investment opportunities.
A surprisingly large number of states and localities have created different forms of family and children's budgets in the last 20 years. We identified over 30 states, counties, and cities that have, at one time or another, produced such a document. (See Appendix A.*1) Such tools are gaining attention and importance in work on child and family well-being at the state and local levels across the country.
The sections which follow address a range of issues about the development and use of such tools, share examples from family and children's budgets that have been produced, and offer what we hope will be practical advice to those considering investing time and energy in this complex but important work.
II. STARTING POINTS
Before tackling the tough questions about how to construct a family and children's budget, let's cover some basics: What is a family and children's budget? Why produce one? How does such a document fit with the larger set of tools necessary to support work on improving results for children and families?
A. What is a Family and Children's Budget?
Perhaps surprisingly, there is not a simple answer to this simple question. Family and children's budgets have taken different forms in different places, covered different territory, and served different purposes. This is as it should be. Our social and political institutions are too complex for any single version of a family and children's budget to be universally applicable. But there are a few basics about creating family and children's budgets that tie together different efforts, and some lessons that can be gleaned from the last dozen or so years of work.
First, a family and children's budget is a supplement to—not a substitute for—existing budget documents. Family and children's budgets add to the set of tools used in the budget decision-making process. They allow us to see in one place a broad array of spending (up to and including all federal, state, local, and private sector spending) for children and families. They can help make sense of our spending for children and families. And, perhaps most importantly, they can be used to steer our strategic use of fiscal and other resources to improve results (*2) for children and families over the long term.
Second, family and children's budgets are developmental efforts. They tend to start out as simple inventories of spending for family and child programs by one level of government, for one or two years. These first efforts usually include only the most basic analysis, showing the proportion of total spending by agency and type of service. Over time, such budgets can grow to include spending from both public and private sectors, and from multiple levels of government, with enough data for presentation of historical spending baselines. They can become documents that contain more sophisticated analyses of trends in spending, of spending across agencies for similar functions, and of the cost-effectiveness of investments in prevention. (Section III below offers a developmental typology for family and children's budgets, which reflects this progression in content, structure, and utility.)
The developmental nature of this work is important in managing expectations. A useful family and children's budget can be created in one year. But it takes more than one year for the budget to become a sophisticated decision-making tool. Public policy-makers should set realistic expectations for first-year family and children's budgets, and press for continued improvement in future budget cycles.
What a family and children's budget is not is also important. It is not a panacea. The mere production of a family and children's budget will not, by itself, change anything. The document must be conceived and developed as a part of the budget process, with information and analyses that are useful to decision-makers.
A family and children's budget is not an accounting tool. It cannot and should not displace the detailed, down-to-the-last-penny budget documents used for appropriation of public funds. It cannot, and should not, substitute for the fiscal-control functions of traditional line-item budgets that are necessary for basic financial accountability. Family and children's budgets should help identify the big-picture, not the little-picture, choices about investment and spending.
And finally, it is not a place to give credit for every last contribution to the well-being of children and families. Creating a useful family and children’s budget document will require decisions about what is included and what is not included, and this may conflict with the impulse to "give credit." The principal purpose of the document is utility for decision-making, and other purposes which diminish this utility should be kept at bay.
Simply put, a family and children's budget is an analytic, policy, and, yes, political tool that can support serious work to improve results for children and families.
B. Why Bother with a Family and Children's Budget?
While there is growing interest in family and children's budgets, producing one can be a lot of work. Are they worth it? We touched on some of the reasons for producing a family and children’s budget in the introduction, but here is a quick summary of pros and cons.
| TO CREATE, OR NOT TO CREATE, A FAMILY AND CHILDREN’S BUDGET
Arguments in Favor
Arguments Against
|
So where do we come out on this? As you might suspect, we here at The Finance Project think that producing a family and children's budget is a splendid idea, providing that:
It is part of a larger tool kit to improve results for children and families. It is a multi-year undertaking, not a one-shot deal. It eventually gets past a Stage I budget to include analyses by function and by result, across public and private sectors, and beyond a single level of government and a single year.
C. Family and Children's Budgets as Part of a Larger Tool Kit
If family and children's budgets are to make any difference, they must be conceived, constructed, and used as part of a larger tool kit and, indeed, as part of a larger strategy to improve results. As isolated documents, they are of limited use. And if they somehow become an end in themselves, they are likely to be a short-lived document of limited use. A Strategy Map for Results-Based Budgeting offers a picture of five essential tools for improving results:
A Results and Indicators List that reflects the conditions of well-being we want for children and families and how we would recognize these conditions in measurable terms.
An Indicators Report that shows how we are doing on the indicators of child and family well-being.
A Family and Children's Budget that shows how resources are used for children and families.
A Cost of Bad Results Report that shows the costs associated with not getting the results we want for children and families (and that provides the financial base for considering potential savings which might be achieved by investing in child and family well-being).
A "What Works" Compendium that makes accessible the successes of others in improving the measurable well-being of children and families.
Creating each of these tools is a developmental effort in its own right. But they can be used together to support a more coherent process for choosing a course of action and aligning the use of resources to support that course.
Other tools figure in this work as well. In addition to the tools listed in the Strategy Map paper, other tools might include performance measurement documents which show how well agencies and their programs deliver service and work to improve the well-being of their client populations ("client results"). Others argue for a range of additional tools necessary to create the "capacity" for changing service systems for children and families. These might include new contracting and program monitoring processes geared to achieving client results; worker-based data systems that support work with children and families across service systems; and tools to support public education and leadership development.
The full development of such a tool set is clearly a multi-year undertaking which some might find daunting. In truth, we have so badly neglected these basic tools for so long that we have some catching up to do. Some have found a family and children’s budget a good place to start because it helps create the partnerships necessary to do this other work. It brings together people around a tangible project in which all have a common interest. The work of creating and using new decision-making tools is, however, parallel and not sequential work. It is not necessary to finish one before going on to another. The Strategy Map paper offers ideas about how to approach the parallel development of these products and processes.
D. Choosing a Common Language (one more time!)
As we get deeper into this business of family and children's budgets, we will be talking more and more about results-based stuff: results-based budgeting, family and children's budgets by "result," etc. What do we mean by "result?" Answering this question requires that we address certain conventions of language that can help us communicate more clearly about this complex work.
There is an astounding lack of discipline in the use of language in the current work on child and family well-being. It is quite common to find people working on these problems who are using the same terms in different, sometimes contradictory, ways, and then wondering why they aren't making any progress.
The following definitions help keep three critical ideas separate, and allow us to communicate more clearly. These are the same definitions used in earlier work about results-based budgeting and decision-making.
Result (or outcome) (*5): A "result" is a bottom-line condition of well-being for children, adults, families, or communities. Results are matters of common sense, above and beyond the jargon of bureaucracy. They are about the fundamental interests of citizens and the fundamental purposes of governments and private institutions. Results are not "owned" by any single agency or system. By definition, they cross over agency and program lines. Results are things such as: children born healthy, children ready for school, children succeeding in school, young people avoiding trouble, stable and self-sufficient families, and safe and supportive communities.
Indicator (or benchmark): An "indicator" is a measure, for which we have data, that helps quantify the achievement of a desired result. Indicators help answer the question: "How would we know a result if we achieved it?" Rates of full immunization help quantify the result, "Healthy Children." Reading scores, math scores, and high school graduation rates help quantify "Children Succeeding in School." And crime rates help quantify "Safe Communities."
Performance measure: A "performance measure" is a measure of how well public or private agencies and programs are working. Typical performance measures address matters of timeliness, effectiveness, and compliance with standards. Performance measures include: the rate of child-abuse investigations initiated within 24 hours of a report, the cost of child-support collections for each dollar collected, and police and fire response times.(*6)
The most important distinction in this set of definitions is between ends and means. Results and indicators have to do with ends. Performance measures and the programs they describe have to do with means. The end we seek is not "better service" (*7) but better results. These distinctions help us describe decision-making and budgeting processes based on clear thinking about what we wish to achieve and how we choose to get there.
III. THE DEVELOPMENTAL NATURE OF FAMILY AND CHILDREN'S BUDGETS
One of the most important and least-understood aspects of family and children's budgeting is the developmental nature of this work. There is a tendency to think that one year ought to be plenty of time to develop a children's budget, and whatever can be completed in this time will suffice. This partly explains why most examples of children's budgets are the relatively undeveloped "Stage I" budgets described below.
There is often considerable pressure to produce a family and children’s budget quickly. This is due in part to the urgent needs of children and families, and in part to the nature of the budget process, which allows only a few windows for exercising influence. A one-year-and-stop approach, however, will leave a family and children's budget largely undeveloped and its utility limited. A children’s and family budget must be built up incrementally over several years.
To help capture this idea of family and children's budgets as developmental entities, we describe three stages in the development process. These stages represent rough groupings of characteristics along the dimensions shown in the chart below. The "defining" characteristic of each stage is the perspective offered on spending. That perspective can evolve from a simple inventory of spending by program, to cross-departmental and cross-sector pictures of spending by function, and, still later, presentation of spending and strategies for improvement by result.
|
|
|
|
Perspective |
Program: Line-item inventory | Functional view across agencies and programs | Results view across systems and sectors |
Sponsorship |
Informal (or outside of government) | One branch of government (Executive or Legislative) | Both branches of government (Executive and Legislative—based in law) |
Scope |
Only one level of
government/private-sector spending (Federal, State, Local, Private) |
Two or More Levels (Federal, State, Local, Private) | All levels of government/private sector spending (Federal, State, Local, Private) |
Time |
Point in time (1 or 2 years) | Historical baseline | Baseline with forecast |
Following is a more complete description of each stage. Appendix F shows excerpts from actual budget documents that include examples from the three stages below.
Stage I: Budget by Program Inventory:
Definition:
A Stage I budget is an aggregation of the program line items associated with spending for children and families as they are represented in the current operating budget. Stage I budgets are usually about the spending of just one level of government and usually exclude private-sector spending, except as it shows up in the form of contracts between government and private-sector agencies. Stage I budgets are sometimes produced by advocacy agencies outside of government.(*8) They most often make use of one to three years of data, drawing on one or more of the following: last year’s actual spending, the current-year appropriation (or estimated actual), and the proposed spending level for the next fiscal year.
Production:
Stage I budgets are the "easiest" to produce, because they mostly involve the use of line-item spending totals already produced in an existing budget document or process. Stage I budgets sometimes start as informal, behind-the-scenes summaries produced by advocacy organizations outside government. However, they can also be more formal official government documents. Stage I budgets involve identifying programs which are wholly devoted to children and families, or the clear shares of programs which serve larger populations. Since they usually involve only one level of government and usually only public-sector expenditures, the problems of duplicative counts are minimal. And such budgets use multi-year data already aligned in an existing document, where comparability between years is not usually a significant problem.
Example(s):
The Kansas Children's Budget, published each year with the Governor's budget submission, is an excellent example of this kind of summary (see Kansas entries in Appendix F). Such budgets have limited, but important, uses in assessing changes in total spending for children. An analysis which compares growth rates in state revenue with growth rates for children and families can answer questions like, "Did children's programs get their fair share of growth in general fund resources?" (*9)
In one state (*10) the family and children's budget was prepared for several years as an unpublished Stage I analysis for use by the Children's Cabinet. It never progressed beyond this form to become a useful tool in the public budget process. Still another state's first attempt at a children's budget was prepared as a simple spreadsheet with no narrative explanation. The document was correctly thought to be too difficult to understand and therefore it was not released. Informal family and children's budgets can pose certain risks if they are not thought of, developed, and presented as the political documents that they are. With no context or explanation, such information can be easily misinterpreted and misused. (*11)
Stage II: Budget by Function:
Definition:
A Stage II budget goes beyond a simple aggregation of existing program line-item spending and presents spending across agency and categorical lines by function. By "function," we mean groupings of related services within the overall family and children's service system, such as the cross-agency set of child care, health care, or community development services, or the more difficult summation of prevention vs. remediation services.(*12) Such budgets begin to account for spending of more than one level of government (federal, state, and local) and spending in both the public and private sectors. And such budgets begin to show historical baselines of spending, not just spending at specific points in time.
Production:
Production of a Stage II budget requires additional data gathering and analysis. Gathering and reconciling data from more than one governmental level involves removing duplicative counts of spending in jointly funded and pass-through programs. As discussed below, "unduplication" is best done by working from the "inside out," that is, starting with a solid analysis of one level and then removing duplicated counts each time the circle is expanded to include another level or sector. Creating functional analyses requires the development of agreed-upon conventions about what functional categories will be used. (See Section VIII for the Los Angeles eight categories, the Oklahoma 11, or the Indianapolis 500.*13) And the development of historical baselines requires both accessing past data, and addressing matters of comparability between fiscal years.
Example(s):
Oklahoma and Los Angeles County show some characteristics of a Stage II budget (see Oklahoma and Los Angeles entries in Appendix F). Oklahoma's budget shows spending for child care, mental health, and other functional categories. This kind of picture makes the budget somewhat more useful because it allows a preliminary assessment of how program expenditures fit together within the service system, and how well they combine to meet needs.
The Los Angeles County Children's budget summarizes spending above the line-item level, classifying expenditures into the functional categories of income support, protective services, health services, juvenile justice, mental health, child care, and prevention. Such functional distinctions can set the stage for efforts to improve coordination of service delivery within these functions. Los Angeles’ classification of program expenditures by functional service area helped advance coordination across county departments by showing areas of related investment and common interest. The most recent work of the Los Angeles County Children’s Planning Council, Laying the Groundwork for Change, Los Angeles County’s First Action Plan for its Children, Youth, and Families (February 1998), is one of the few documents that links three of the key tools in results-based budgeting: the children’s budget, the indicator report, and an action plan of "what works" strategies.
San Diego's Future Scan, a one-time effort produced in 1993, shows one of the broadest pictures of functional spending across the federal, state, local, and private sectors of any family and children's budget. The San Diego entries in Appendix F show spending separated by public-sector government, private sector, and education. Public and private sector funding is further disaggregated by fund source (federal, state, county, city, and private).
Functional classifications also become useful in creating an investment case for children's spending as discussed in the "cost of bad results" section below. (*14) Colorado’s Children’s Investment Prospectus (1997) uses six functional categories to advance the potential benefits of investing in the well-being of children ages six and under.
Stage III: Budget by Result:
Definition:
Stage III budgets are the final frontier in family and children's budgeting. A Stage III family and children's budget builds on the previous two stages and provides not just program and functional pictures but a results-based view of expenditures as well.
We are just beginning to understand what a results-based budget document looks like. Such documents could have separate sections or volumes devoted to different perspectives. Volume 1 would present the highest-level view by result, across agencies and across the community. The document would present indicator baselines that describe the extent to which each result is being achieved, along with a summary of current strategies to improve results, including the work of both public and private partners. Volume 2 would present the program components of these strategies and provide information on the performance of each. These volumes together provide for both cross-agency accountability for results and within agency accountability for performance. (*15) (See the Results-based Budget Schematic at the end of Appendix F.)
Production:
Production of a results-based family and children's budget is not just a matter of reshuffling and recategorizing items in the line-item or functional stages. Programs may be presented as part of more than one result for which they make up an important strategic component. This means that, unlike line-item and functional summaries, the results summaries may involve counting a particular program in more than one place. The production challenges in a Stage III budget involve data-gathering on spending and investment across all levels of government across public and private sectors, consensus forecasts of spending trends, and politically useful ways to present strategic choices by result.
Example(s):
There are no existing examples of fully developed Stage III family and children's budgets. But there are some efforts that shed light on what such documents might look like.
The Contra Costa County, California, Children and Family Services Budget for 1997-98 shows some characteristics of a Stage III budget. It includes both a functional summary of spending and a set of charts linking county programs to results and the costs of bad results.
In Fiscal Year 1995, the budget for Multnomah County, Oregon, began showing the relationship of the county agencies to the County's urgent benchmarks, and provided a summary of both ongoing and new efforts to address each "urgent" benchmark.
Vermont's Agency for Human Services budget for FY 1998 incorporates some analysis of spending for results across the human service and education systems. These efforts are the precursors of what results-based budgets and results-based family and children's budgets will look like in years to come.
IV. HOW DO YOU BUILD A FAMILY AND CHILDREN'S BUDGET?
Issues of Content and Construction
We here at FPL (Finance Project Laboratories) have gathered as many family and children's budgets as we could find. And we have used the most modern scientific methods to extract lessons from these budgets. This has led to significant advances in our understanding of family and children's budgets, and most importantly, a list of construction issues that you may wish to consider in creating your own family and children's budget.
A. What Do We Mean by Children and Families?
This sounds like a simple question, but it is not.
Defining "children" (*16)
There are many differences across the states in the legal definition of child, with age range (0-18 or 0-21) being the most important. In some cases, programs — like welfare, social security, and special education — consider recipients of benefits to be children up to age 21 if they are in school or training, and up to age 22 if disabled. Conversely, there is a growing and controversial trend to count children below 18 who have committed certain serious crimes as adults in the criminal justice system.
Defining "families"
The matter of defining families is even more complex than children. We are a society of many different kinds of families, and simple definitions don't work. In some benefit programs (such as TANF and Food Stamps), the definition of family (or household) is tied directly to eligibility, and the definition can get quite complicated, dealing with varying degrees of relationship and combinations of living arrangements. Aside from eligibility, defining families can be a politically charged undertaking.
Allow multiple definitions to coexist
There are a few simple things to keep in mind here to keep from getting lost in this definition and data forest. First, be practical. Don't get bogged down in trying to craft a perfect set of definitions of children or families. The nature of this work requires some ambiguity about this, and you might as well get used to that at the start. It will, in fact, be necessary to allow differing, even contradictory, definitions to coexist. The simple rule of thumb is to count expenditures for children and families using the definition of the program in question. It is not necessary (and, in fact, it is not possible) to reconcile differing definitions across programs. If you use common sense about matters of definition, the product will be OK.
Opt for inclusion
A more important issue has to do with whether we include all children and families or just some children and families. Some family and children budget efforts have taken as their subject not all children, but "at risk" children or children with special needs. Others have argued that we can and should leave out certain kinds of expenditures, such as elementary and secondary education, because they are so large they will dwarf other expenditures or will give the impression that we are already spending "too much" on children (see the backlash discussion below).
These are not technical arguments about what we know, or what we can produce. They are political arguments about the purposes and uses of family and children's budgets. We do not impugn the motives of those who advocate such positions. But we argue for inclusion, for two reasons. First, credibility. Everybody knows that education spending is about children. To leave it (or any major category of spending) out of such a budget simply detracts from the credibility of the product. If readers/users of family and children's budgets feel that they are being manipulated by the deliberate exclusion of important information, then the document won't be credible and it won't be used. (As noted below, education spending can be treated separately within the budget. It is often helpful to show totals with and without education spending.)
Second, when we choose to count only programs for "at risk" children, or "low income" children, or "disadvantaged" children, we further the kind of distinctions about children that have gotten us into trouble in the first place. These distinctions, and the terms we use to support them, reinforce categorical thinking about children and lend credence to the divisive idea that the "problem" with children is about "somebody else's children." Family and children’s budgets can serve to advance the notion that we have a stake in the well-being of all children.
Give special, not exclusive, attention to sub-populations
Within a broadly constructed family and children's budget, there is good reason to give special attention to certain sub-populations of families or children. If we intend to improve overall results for children and families, then special efforts will be required for children with greater needs, such as children in state custody, children in special education, and children in low-income families.
It is appropriate and helpful to use the "platform" of a family and children's budget to report on the well-being of these children and to assess the adequacy of efforts to improve their well-being. This can take the form of special breakouts of data by sub-population and special sections of the budget document devoted to recommendations and action plans. However, it is best to think of such sections and analyses as supplements to, not substitutes for, a broadly based budget for all families and children.
B. The Basics of "What’s in?" and "What's out?"
Given an answer to what is meant by children and families, the next question involves what programs and expenditures are to be included. Since a family and children's budget is a summary of the financial (and other) resources devoted to families and children, we need a good working definition of "resources devoted to families and children."
In truth, just about all governmental expenditures can be connected to children in some way. The transportation department builds roads on which children travel. NASA produces pictures of Mars that children watch on television. The defense department protects children along with shipping lanes. So where are the useful boundaries for a budget about children and their families? There are several parts to the answer to this question.
Utility bottom line
Going back to first purposes, family and children's budgets are intended to create a more complete and coherent picture of spending for families and children, so that we can make better decisions about the way we spend money, structure services, and invest in well-being. So it makes sense to have a dividing line between what will help do that, and what will not. Including a part of the NASA or the roads budget will generally not help.
A sorting process
It is useful to think about this work as a sorting process that makes use of three large categories: (1) Things that are definitely in; (2) things that are definitely out; and (3) things that could go either way. The reason that this seemingly common-sense (but often missed) approach is so important is that it allows the parties involved to reach agreement quickly on what is definitely in and definitely out, and then to concentrate their discussions on the items in the middle.
It is essential to understand that decisions made in this part of the process are not carved-in-stone matters of "right and wrong." Such decisions may be revisited over time, and they may be refined as the development process proceeds. What is "in and out" is as much a political judgment as a factual judgment. The test, again, is utility. Does the information help us make better decisions about spending for children and families? If yes, include it. If no, don't.
Following is a general look at each of these categories:
Pretty Definitely In:
Expenditures which directly benefit children, or that benefit children through investments in their families.
These expenditures include direct spending such as elementary and secondary education; child care and early childhood education; child welfare and juvenile justice services; income supports for families with children (such as TANF, Supplemental Security Income (SSI), and Food Stamps); Medicaid, Child Health Insurance Program Funding, and other spending for medical care; and housing subsidies and supports.
This category also includes indirect spending on community building and community development, in other words, investments that bear on the quantity and quality of community supports for families raising children (e.g., economic development targeted to families with children, business development supporting families like child care facility funds), and community development initiatives such as family centers and playgrounds.
Pretty Definitely Out:
Expenditures for infrastructure that benefit all members of the community more or less equally.
These expenditures include such things as roads, bridges, sewer and water services, environmental protection, etc. It is generally not going to help us make better decisions about investing in children and families to tally spending on infrastructure.
To be Decided:
Many parts of government which serve the general population have specific components that relate directly to children and families. Although police departments might appear to be "pretty definitely out" at first, they often have special units devoted to child abuse or domestic violence. There may be an important benefit in asking police what portion of their time they spend on youth. One county participant noted, "...you ask how many of their arrests are youth, and the light begins to dawn. They’ve been into the geography of precincts so long that the age break rarely occurs to them as a budget tool. We broke out the percentage of patrol that is devoted to youth suppression and arrest, community policing dedicated to youth, etc., and it is a 20% base of a very large budget."
Most state and local court systems have special coverage of family and juveniles matters. And family courts are becoming more common. In these cases, it is possible and desirable to allocate a portion of the cost of these services.
For many other services, it is possible, but probably not desirable, to allocate a portion of expenditures to children and families. A family and children's share of expenditures for fire and rescue services could be identified, based on the percentage of households with children in the service catchment area. But does including this information help us improve the conditions of well-being for children? One could make the case that fire safety should figure in our work on child safety, and that the fire department therefore has a useful role to play. But is it necessary for a portion of the fire department's budget to be in the family and children's budget for this connection to be made? Probably not.
One way to take the edge off the decision about what to include or exclude is to write about the decision process in a section of the budget document itself. It is possible to list those services, supports, capital investments, etc., that indirectly benefit children and families for which detailed expenditure information is not included. This allows some of the important connections to be acknowledged without burdening the budget document itself with information of questionable utility.
C. Using Rule Sets for What's In and What's Out
A second, more disciplined, way to look at the question of "what's in and what's out" is through "rule sets" that help guide the process. (See Appendix D.) While there are no hard-and-fast rules, we have learned some things from looking at many different children's budgets, by observing the process, and by actually doing this work ourselves in constructing a database of federal spending for children.
Consider the following summary of two rule sets as a starting point.
Basic rule set
This rule set envisions a three-step sorting process that identifies public-interest expenditures benefiting children and families with children.
Step 1:
Separate "public interest" from other spending: Include spending of federal, state, local, or private non-profit funds which serve interests relevant to the public life of the community as a whole. Exclude private, for-profit spending, and personal spending by individuals or families for their children.
Step 2:
Separate services that benefit families and children from universal services: Does the public-interest spending in question constitute a universal service that benefits all citizens more or less equally? If yes, exclude such services. If the spending addresses children or families with children in some unique way, then include it.
Step 3:
Identify the full amount or apportioned share that benefits families and children: For the remaining expenditures, include the full amount if the spending is fully attributable to families and children. For other expenditures, include an apportioned family and children’s share. Apportion on the basis of client/customer population share or, where available, the proportion of expenditures devoted to children and families.
Other sorting rules and categories
Database software is sometimes used as a tool in the construction of a family and children's budget. When it is used, the array of data fields, definitions, and codes used in database construction constitutes a rule set for categorizing expenditures.
A more complex set of decision rules is illustrated by the work of The Finance Project in developing a database of 1994 federal spending for children, families, and communities. This work required that federal spending across many agencies and programs be identified and differentiated on the basis of its relevance to the well-being of children and the families and communities in which they live.
The database included three data sets:
Investments in child well-being (programs that directly benefit children); Investments in family well-being (programs that directly benefit families with children); and Investments in community well-being (programs that indirectly benefit children, and families with children, by strengthening the communities in which they live).
Separating these three categories served to ease data collection, sorting, and apportionment, and allowed greater flexibility at the time of analysis. Expenditures could be combined from any of one, two, or three data sets, depending on the type, purpose, and scope of analysis.
The data-entry process also allowed for identification of other distinguishing program characteristics, such as:
Target eligibility: Who is eligible for the program or service? What special rules apply to individual eligibility? (i.e., means-tested or special needs eligibility); Type of service: What type of service is provided? (e.g., education, health, literacy etc.); and Program function: What is the primary function of the service? (e.g., primary prevention, remediation, training, etc.).
Information of this nature helps to specify how dollars are intended to reach their beneficiary populations. For each program, the database also included information on the administering agency, the funding mechanism, fund allocation formula, and fund match requirements, if any. A complete set of data entry formats is provided in Appendix B.
D. The Importance of Community Investments
A family and children's budget should also include those resources invested in community well-being which indirectly benefit children and families. It is increasingly clear that the characteristics of the neighborhood/community environment in which children live is critically important to their chances of healthy and successful growth.
In 1990, 17 percent of children under age 15 in the nation's 50 largest cities lived in "distressed" neighborhoods (defined as communities with high concentrations of poverty, female-headed families, unemployment, and welfare dependency). Economic and community development investments may be among the most important investments we can make in children and families. These kinds of investments can and should have a prominent place in family and children's budgets.
But community investments, like infrastructure, are usually investments that span beyond families with children. In some cases these can be directly attributed to families and children, as in the case of low-income housing for families with children, job development targeted to families on welfare, or school-to-work transition. Community development components like development of parks, playgrounds, recreation centers, and family centers are also closely tied to family and child well-being, and may fit cleanly within the inclusion categories noted above.
But other economic and community development efforts -- such as those targeted at more general business development, safety, or strengthening of community institutions -- are vitally important to the well-being of children and families, but are not exclusively about children and families. In these cases, it is not useful to allocate a share of such investments in the budget document.
As an alternative to allocating pieces of community building and development, Stage I budgets could inventory such investments where they occur (e.g., Empowerment/Enterprise grants, Economic and Community Development grants). In Stage II budgets, it would then be possible to show these important efforts as a separate "economic and community development" function, presenting an analysis of investments in communities, parallel to similar presentations for other functional areas (such as income supports, child protection, child care and early childhood education, etc). Stage III budgets could then include sections that focus on turning the curve on indicators of community well-being. These investments will figure prominently in the action agenda and budget components of such sections.
The matter of identifying spending on community well-being is, or can be, closely linked to the work of community "asset mapping," as advanced by John McKnight, or "Youth Mapping," developed by the Academy for Educational Development and others. The idea of approaching communities as places with assets to be preserved and enhanced, not deficits to be remedied, is a powerful and compelling idea. And it changes not only what is counted, but also how it is presented and used. While this brief reference does not do justice to this important body of work, there are significant benefits to linking these efforts, particularly at the county, city, and neighborhood level.
Whatever approach is taken, the evidence about successful change strategies (from Empowerment Zones to Community Policing) suggests that the economic and community development perspective is vitally important to successful efforts to improve results for children and families. And family and children's budgets, at each stage of development, should place this perspective front and center, along with the more traditional definitions of programs associated with child and family well-being. This kind of analysis is now largely missing from the work on family and children's budgets, and represents one of the most important "frontiers" for development in the future.
E. The Matter of Private-Sector Expenditures
The same "definitely in, definitely out" rules generally work with the private sector, but the "to be decided" category can get somewhat more complicated. This is due in part to the diverse array of organized private-sector spending—from charities, philanthropies, non-profit service providers, and volunteer, civic, and business organizations — and the largely unquantified expenditures families make directly for their children.
Two criteria, one of which is discussed above, will generally serve to sort things out. Is the spending a matter of "public interest" or public policy? And, is the spending "material"? Materiality is a principle of accounting and, sometimes, law. Is the spending significant? Does it amount to anything? Can it be left out without affecting anything important?
Not everything will split cleanly along these lines, but the basic principle of identifying "material, public interest" expenditures should help guide the work. United Way spending on recreation is an unambiguous public-interest expenditure on behalf of children, and in most communities is large enough to warrant inclusion. The sponsorship of children’s sports teams by local businesses is less clearly a matter of public interest, and, in any event, is generally small enough to leave out.
The matter of parental spending on children deserves some special attention. Parental spending on children actually makes up one of the largest categories of total spending. This includes such spending as basic food and shelter, recreation, family shares of health care, and education. A U.S. Department of Agriculture study estimated the cost of raising a child, born in 1996, from birth through age 17 at $241,440 for middle-income families. But does information on parental spending constitute a matter of public-interest spending? Can it be separated from clear public-interest matters associated with the adequacy of family income? Does information on parental spending belong in family and children's budgets?
As a practical matter, it is probably best to exclude parental spending from budget totals, unless there is a clear relationship between such spending and public policy (such as the co-payment shares of child care, health insurance, or other services). A decision to exclude parental spending should not be construed as a judgment that such spending is somehow unimportant. Good data on such spending is extraordinarily difficult to get. And this issue could easily bog down the development process. Private parental costs can be referenced in the budget document, and available information presented as part of the analysis, not budget, sections.
The place where private spending by families and public policy clearly intersect is in the area of tax policies designed to supplement and/or encourage private family spending for children. As noted below, tax expenditures, such as child care subsidies, and earned income tax credits belong in the family and children's budget. It generally makes sense to include the public, but not the private, share of such expenditures where they occur.
V. DATA CHALLENGES IN BUILDING A FAMILY AND CHILDREN’S BUDGET
We have described the three stages of family and child budget development and addressed the basic process of sorting what's in and what's out. But actually building family and children's budgets is about data. And even where it is possible to clearly identify what to count, gathering the data is another matter. The following sections address some of the most important data challenges. Don't be discouraged by this list. It is important to start with what you have and improve over time.
A. Geographic Boundaries
In this country, we have a paper-mache form of government. The overlapping governing structures of federal, state, county, city, and school district sacrifice clarity of accountability for overall strength of the system. But they make for a nightmare when you are trying to gather data for a particular area. Some of the biggest disconnects:
Federal spending data (SSI, for example) is usually available by state, but may be difficult to get by county, city, or other jurisdiction. When such data are organized by zip code or census track, they can be summed to higher levels, but often only with considerable effort and expense. Where county data can be obtained, they are often not broken out by cities, towns, or school districts inside the county. Health and human service expenditures, for example, are often available at the county level, but are difficult to get at sub-county levels (say, for a city children's budget). School-district boundaries often do not match county or city boundaries. There are about 15,000 school districts in the United States and about 3,000 counties. Boundaries match in only a few states.
Where possible, family and children's budgets should be constructed with "clean" geographic boundaries, for the simple reason that public decision-making bodies are organized by geography. If geographic boundaries are not clear, then accountability for what the budget shows is clouded, and decision-makers who might otherwise use the document will be given a reason not to.
One interesting solution to this problem is illustrated by the indicators report produced for local school districts in Vermont (produced annually by the Vermont Agency for Human Services in collaboration with the Vermont Department of Education). The report shows trend information on education, health, and social service data at the school-district level where this information is available or can be produced. Where the data cannot be split out by school district, the next-higher level of aggregation (either county or regional grouping of school districts) is shown. This allows readers to see the data for their specific school district or the next larger area of which they are a part. Both views allow for some "ownership" of the data by policy-makers and citizens alike. Although this report is not about expenditures, the same principle can be applied to expenditure presentations.
B. Double Counting
It is quite common for the same money to show up in more than one budget. This is almost always the result of joint or pass-through funding between levels of government, or contracting between agencies of government and private-sector agencies. The extensive use of Medicaid to fund a wide range of health-related services across state and local agencies is one of the most complex double-counting problems.
Several family and children's budgets have dealt with double-counting problems by showing total spending where it is budgeted and then taking out the estimated double counting from the bottom line totals for the entire budget. The Utah Children and Youth At Risk Budget uses this method and provides a detailed estimate of "duplications" mostly attributable to revenue transfers.
There are several choices to be made here. First, is it possible to unduplicate expenditures and is it worth it? Generally, unduplication is worth the effort, but it is almost never possible to do this completely. Adjustments and estimated adjustments are OK.
Second, if we can identify double counting, should we reduce the duplicated amount from the fund source agency or the fund use agency? As a general rule, the expenditure should be left in the fund use agency and reduced from the fund source agency. This is for the simple reason that for most analyses, fund use is more important than fund source in showing what kinds of services are provided to whom.
For example, consider the widespread practice of using Medicaid funds to support special education services. If these expenditures were removed from the special education budget and shown only in the Medicaid budget, we would lose the picture of total spending on special education. And special education use will be lost in the vast array of other services supported by Medicaid. It is better to show the expenditure in Special Education and to show Medicaid as one of the fund sources within Special Education. In other words, the fund source analysis is subordinate to the fund use analysis. The display of fund sources for a given program or function will typically show the split of federal, state, and local funding, and may also show individual fund sources such as Medicaid. In this way, both fund use and fund source pictures can be presented without distorting the more important fund use analysis.
One process that helps address this problem is to construct the budget from the inside out. That is, start with spending information from the smallest level of government to be considered in the budget construction process. Then successively add information from larger levels. This convention allows the best use of summary reports on [federal to state; state to county; county to city grants and transfers], which can be subtracted as part of the unduplication process.
A more complete solution to the double-counting problem requires the creation of new reporting of disaggregated spending by jurisdiction and program. The current state of most automated government systems makes this type of analysis difficult, if not impossible. But as such data systems improve, the task may fall within reach.
C. Splitting Program Expenditures
In many, if not most, cases, expenditures for children and families will break cleanly by program. That is, whole programs and their expenditures can be counted in the family and children’s budget, without adjustment or modification. But many programs, important to family and child well-being, serve other populations as well. Medicaid, Food Stamps, and SSI, for example, provide assistance to elders, single adults, and childless couples. The education system serves adults without children in adult education and vocational rehabilitation programs. And higher-education expenditures clearly include both children and adults. In some programs, it can be difficult to separate child and family expenditures from other expenditures. This is sometimes a thorny problem with no easy answers. But a few simple guidelines may help.
First, let the program providing the information make the split whenever possible. This cuts down on the work of producing the family and child budget document, and adds to the credibility of the data. Make sure, however, that the program provides information on the methodology used, so that you can make judgments and answer questions about the resulting numbers.
There are a number of methodological choices when it comes to splitting program expenditures. It is best if there are data systems within the program which separate expenditures for services to children and families with children from other expenditures. Where this is not the case, total program expenditures can be allocated based on one of several percentages:
Percentage of clients, students, or customers who are children and parents of children, or Percentage of dollars spent for children and parents of children.
Of these two, it is much better to use the dollar percentage, if it is available. It is often the case that children represent a disproportionate use of resources in a program. Sometimes, this difference is dramatic. In Medicaid, for example, children and parents of children make up about two-thirds of the total client population, but only about one-third of program cost. Expensive long-term care costs for the elderly and disabled adult populations are responsible for this difference.
Where program data are not available, it is sometimes possible to split programs on the basis of the percentage of children and parents in the general population, or to use splits which have been established in other jurisdictions with better data. These are, of course, last resorts. Whatever method is used, it is important to carefully record the approach in the notes to the budget document, so that readers can make their own judgments about the data, and future budgets can improve on past work.
D. Time Boundaries Part 1: Fiscal Years
Just as there are different geographical boundaries, there are different time boundaries used by different organizations. The federal fiscal year runs from October to September. Forty-six of the 50 states and most local governments operate on a July-to-June fiscal year. And many non-profits operate on a January-to-December fiscal year. School districts operate on a school year, which usually, but not always, corresponds with local government fiscal years.
Producing a family and children's budget for 199X could mean any one of these (calendar 199X, or fiscal 199X by any one of four possible definitions). There is actually a fairly simple solution to this problem, but it requires some (Zen-like) compromise in precision: Take whatever is available, and treat calendar and fiscal 199X as the same for aggregation, display, and analysis purposes. The alternative to this solution is to try to allocate expenditures to a uniform time period. This can be done, but you will drive yourself crazy doing it.
E. Time Boundaries Part 2: Baselines, Trends, and Discontinuities
Most family and children's budgets are point-in-time budgets. But the most important questions we need to ask are about spending trends. And, of course, there can be no trend data without more than one year—and two years isn't much better. Take, for example, the children's budget effort in one state that compared spending for two points, 10 years apart. The analysis showed dramatic shifts in spending patterns toward remedial costs and away from prevention. This is clear and useful information, but the more recent trend information, particularly at the program or function level, would add considerably to this picture. Did expenditures peak during this period and begin to decline, or did they bottom out and begin to increase?
Generally, we need budgets to go back three to five years, that is, show three to five years of actual expenditure history. This problem, of course, takes care of itself, prospectively after a few years. But the utility of a family and children's budget is greatly enhanced if you can get multi-year data from the start.
One important limitation in collecting multiple years' data is comparability across fiscal years. Not only do programs and organizations change, but definitions of data and budget categories change within existing programs. This can make comparability of data a problem even within the same program and the same jurisdiction.
There are two responses to this problem—one is usually right, the other is usually wrong. The "usually wrong" response is to try to adjust the data so that they are comparable. This can sometimes be done where programs have simply been renamed, or where program components have been realigned in a clear way that allows an easy crosswalk between the old and new system. But usually more complex attempts at making the data comparable are necessary, and they are generally not worth the effort. The "usually right" way to deal with this comparability issue is to treat it as an analytic, not data problem. Allow the different definitions to coexist in the presentation of spending totals by agency, program, and function. And then consider any changes in program policy, budget alignment, and data definition as part of the process of analyzing and presenting multi-year trends. (See the menagerie of analyses in Section B below.)
One other lesson relates to the matter of multi-year data. It is generally a good idea to gather as many years worth of data as you can for a given program at one time—for the simple reason that it will be easier than if you have to return to the task later. Gathering multi-year data in one pass means that you need to open source documents fewer times, and thus you will have an easier time identifying and understanding comparability across fiscal years. Most annual budgets present three years’ worth of data (actual data for the past budget year, the current budget year appropriation, and the proposals or approved budget for the next year). Biennial budgets typically present four to six years of data. Multi-year data is essential for analysis of spending trends, one of the most important types of analysis that can be developed with family and children's budgets.
F. Funding within the Public Sector
Within the public sector, there are many fund sources that support services to families and children. It is useful to have at least a basic split of federal, state, and local funds. And more detailed funding information by specific fund source will be useful in later stages of work. Such information allows important analysis of spending by fund source:
Assessment of the fiscal stakes of current policy for various funders. What is the county’s general fund stake in spending for the well-being of children and families? How much does the federal government contribute to services for children and families? How much of this is for remediating problems? How much is for preventing problems? Analyses of cost shifts between funders over time. Such analyses can shed light on the well-known shifts of costs from the federal to the state level (such as those that result from cuts in block-grant funding) or shifts from state to county government resulting from realignment or other changes in state law governing joint funding responsibilities. Funding information by fund source also may offer opportunities for refinancing and revenue maximization, when current fund combinations are not the most advantageous ones available.
Most important here is the separation of "general funds" from other fund sources. General funds are the funds over which state and local decision-makers have the most discretion. The utility of the family and children's budget as a decision-making tool will tie closely to its ability to inform or illuminate general fund decisions in the budget process.
G. Funding within the Private Sector
Getting anything near complete information on private-sector spending is one of the most difficult challenges in constructing a family and children's budget. This is due in part to the many different kinds of private organizations that exist in a given state or locality. But, more importantly, it is due to the fact that there is no place where all of this information is routinely brought together.
The easiest way to deal with this is to build on the work of one or more umbrella organizations that support services for children and families. Organizations such as the United Way can be valuable partners in a family and children's budget effort. And in some places, like San Diego, the United Way has played a leading role in developing such budgets.
It is also possible to conduct a special survey of the agencies in your area or to create a process for regular reporting. This approach has been used successfully in Los Angeles. But there is an important caution to be raised if you intend to do this: Make the best use of existing information before you ask people for new reporting. And keep new reporting as simple as possible. If you create complex new reporting requirements for already overburdened private organizations, you won't get what you want anyway, and you'll create a lot of resentment in the process.
Finally, it is possible to use the results from somewhere else to estimate private-sector spending. San Diego used the results of the Los Angeles survey to estimate San Diego private-sector spending. There are also rough national estimates available in the Statistical Abstract of the United States and from other national organizations. (See Appendix D.)
H. The Matter of Tax Expenditures
Another important way in which money is "spent" on children and families is through the tax system. Tax credits or tax deductions are used by both federal and state governments to provide direct financial benefits to families with children. Among the most important of these:
The dependent care deduction, part of all federal and state income tax calculations. The federal child care tax credit. Twenty-two states also provide a state tax credit, usually calculated as a share of the federal credit, and four states provide a separate state income tax deduction. The federal Earned Income Tax Credit (EITC) is one of the most important income transfer provisions for low-income working families. In 1994, 17.2 million families received approximately $18.7 billion. A number of states have also enacted state-only EITC provisions, including New York and Wisconsin. The $400-per-child tax credit and special tax allowances for tuition, enacted in 1997.
Tax expenditures for children and families belong in family and children's budgets, but may require special explanation, and should be carefully separated from traditional expenditure information.
I. The Matter of Capital Expenditures
Capital funds are often overlooked when it comes to identifying funding for children and their families. But capital expenditures play an important role in the overall financing of services for children and families.
The single most important child and family capital expenditure is school construction and repair. In 1996 the General Accounting Office found that "about one-third of the schools nationwide...reported at least one entire building...in need of extensive repair or replacement." "Moreover, about 60 percent of schools nationwide...reported needing extensive repair, overhaul, or replacement of at least one major building feature..." The GAO report went on to estimate "that schools nationwide needed to spend about $112 billion to repair or upgrade them into good overall condition."
Other capital expenditures related to children and family services are also important. Some states have used capital funds to support start-up costs for child care facilities or homeless shelters. And even the basic capital costs of adequate office space for family and children’s services can be important to track. In addition, the capital budget is a sometimes forgotten part of the budget process where new resources can be "won" for children and families. Family and children’s budgets should include a capital funding summary as a separate section or as part of the summaries by type of service or function.
J. The Matter of Revenue
There are also times when revenue is directly connected to children and families. The clearest examples are the Children's Trust Funds that have been set up in many states. These are often funded through tax check-offs or other special revenue provisions. Other revenue also may be dedicated to children and families. Examples include:
Florida, which has special taxing districts in three counties that raise funds from a percentage of the property-tax levy and use the funds to finance services for children. California, which devotes a percent of the tobacco tax to smoking/drug-use prevention for young people. San Francisco, where Proposition J puts money aside for investments in youth. Massachusetts, in which money was raised for children's services as a 5% "rider" on a revenue bond for prison construction. In some places, fees are devoted to special purposes, such as the use of marriage-license fees for domestic-violence services.
Raising the issue of special-purpose revenue is not intended as an argument for or against this approach to revenue generation. But jurisdictions with these types of provisions will need a section of their family and children’s budget devoted to revenue matters.
K. The Matter of Inflation and Population Growth
It is important to take account of inflation and population growth (or decline) in analyzing and presenting family and children's expenditures. This is typically done with the Consumer Price Index and population totals from Census actual or estimated data. The reason is simple. If spending is growing more slowly than the combined effects of population and inflation, then it is actually declining in real or "constant" dollars. Analyses of spending trends are often most meaningful when presented in terms of "constant dollars" or "constant dollars per child."
There are a number of technical challenges with this kind of analysis that we will touch on only briefly here. First, the CPI is a blunt instrument when it comes to inflation adjustments for family and children's services. There are three big components of spending on family and children’s services that do not track well with the overall CPI: (1) Medicaid expenditures and other medical costs have been rising faster than general inflation for many years. While this rate of growth has moderated, it still does not match the overall CPI. (2) For the salary component of education and other public and private services, cost growth is often tied to salary increases set in the public budget process, which are often unrelated to actual cost of living changes. And, (3) the components of government spending for the purchase of supplies and service bear no relation to the market basket used to compile the CPI. It is quite common to find that these components of family and children's services grow much faster or much slower than the CPI as a whole, thereby seriously distorting a CPI-based analysis. It is possible, and probably desirable, to use the separate medical and "government consumption expenditures" price indices as adjusters instead of the CPI.
L. The Matter of Software
Last, but certainly not least, among the technical questions is what kind of software to use to support this effort. Family and children's budgets involve a forest of information, and any effort beyond the most basic inventory will quickly involve the construction of a very large database. The "simple" answer is to use one of the many software packages that has data base, spreadsheet, and graphics capabilities. The database components will be most useful in gathering data. Database software will allow the data-gathering effort to add data fields over time as the work becomes more inclusive in later stages of development. The spreadsheet and graphics components will be most useful in analyzing and presenting the data. In addition, there will sometimes be a need for statistical packages for analytic and forecasting work. And finally, Internet formatting and presentation software will enhance access and distribution.
VI. WHAT DO YOU DO WITH A FAMILY AND CHILDREN'S BUDGET?
Issues of Use
The most important question to answer in the design and construction of a family and children's budget is, "What in the world are we going to do with it once we have it?" Often, this question gets asked too late, after all the work has been done to gather and present the data. Then you discover that the most important questions you want to answer can't be answered with the data you have.
So ask this question first. Imagine that you had the family and children's budget of your dreams already completed. What kind of analysis would you do? What questions would you try to answer with this new tool? What audience would you try to address? What specific kinds of arguments (e.g., invest in prevention, improve coordination, fill critical gaps, etc.) would you try to craft? What would be the most powerful kind of presentation you could make to support this position?
The intent of this section is to stimulate thinking about the answers to these questions, not to show every conceivable type of analysis that can be done. Appendix F shows some of the best charts, graphs, tables, and other presentation formats actually used in existing family and children's budgets.
The following section presents some ideas about the kinds of analyses that can be created from increasingly sophisticated and increasingly complete family and children's budgets. This presentation is not exhaustive, although it may be exhausting. The progression is cumulative, that is, analyses that can be completed with Stage I budgets will usually also be contained in succeeding stages of development.
Appendix F contains copies of all the analyses referenced in the text. Appendix F is organized in alphabetical order by state, and by county/city within state.
A. A Menagerie of Analyses for Stage I Budgets
Spending by agency, type of service, and fund source
What are the relative proportions of spending by agency, type of service, and fund source? Stage I family and children's budgets almost always include a table or pie chart showing these proportions, since this is one of the relatively few things you can do with a Stage I budget. But these charts don't tell us much that is surprising. They tell us which agency is largest, which service is most expensive, which fund source pays the most. When education expenditures are included, they are (surprise!) the largest (usually more than 50%) part of total spending.
Proportional agency, service, and fund source spending charts give us a rough sense of the shape of spending, and can be helpful for educating a public accustomed to thinking that cash welfare payments are the largest portion of total government spending on families and children. And these kinds of presentations can shed light on the many different agencies, organizations, and funders with a financial stake in the well-being of families and children.
Pie charts and other proportional displays can also help explain why some funding issues are more important than others in a particular jurisdiction. In many California counties, for example, the largest single category of county spending (net county cost) for families and children is juvenile justice. At the state level, education is the largest category. These differences are important indicators of incentives and pressure points in the budget process.
Fair share of growth
Are families and children receiving their "fair share" of growth in resources? In times of cuts, are they protected more, or less, than in other parts of the budget? These are among the most basic questions that can be answered by a Stage I family and children's budget.
The analysis is usually done separately for Total Funds and General Funds. Total Funds are all the fund types included in the budget, and they encompass federal, state, local, special, and other funds. The total-fund analysis tells whether total-fund resources available for family and children's programs are increasing or decreasing.
However, the rate of growth question for general funds is often more important than the total fund analyses. General Funds, for most levels of government, are the principal form of discretionary money. Decisions about general funds are, therefore, often the "real decisions" in the budget process. Total-fund analyses sometimes reflect matters wholly beyond the control of the state or local jurisdiction. Increases, or cuts, in federal funds can sometimes be the overwhelming determinant of whether total spending goes up or down for many family and children’s programs. Economic and demographic factors can dramatically affect total-fund spending for programs, such as welfare. General-fund spending, however, is more directly under the control of the state or locality in question. And general-fund analyses can tell the extent to which decision-makers are putting their money where their mouths are.
General funds are not the only source of discretionary money, however, and fair-share analyses can be designed to look at other discretionary fund sources as well. These include federal sources such as capped federal block grants (e.g., Social Services Block Grant, Child Care Block Grant, etc.), and special trust funds that are sometimes financed with receipts from tax check-offs or fees. In one state, these fund sources were combined with general funds to create a category called "General Fund Equivalent (GFE)" used for assessing the total set of fungible discretionary resources.
The analysis of general funds fair share can be illustrated by work done following publication of the 1993 Kansas Children's Budget. (See Appendix F.) This analysis showed that state general funds increased by 4.4%, while children's spending increased by only 1.2%. The fair-share analysis posed the simple question, "Why did children’s programs receive less than the average growth of all state government?" This analysis was used to press the case for more investments in children in the budget deliberation process.
An important, and surprising, point is relevant here — raised by a state Senator, following a presentation to the Budget and Tax Committee in Maryland. "You know," he said, "It's not always a good thing for spending on children to go up. Don't we want spending on things like juvenile crime and foster care to go down?" The answer is yes. The section below on the cost of bad results addresses this perspective. In fact, it would be a good thing if spending for bad results went down, so that there would be more money to spend on matters vital to healthy child and youth development — like education, child care, and health insurance for children. See Michigan's analysis, Table 1 in Appendix F. This table presents general fund spending by service type for FY 1991 and FY 1996. Incarceration of youth in adult prisons went up by 121% (an area where growth is not desirable). Early childhood education services went up by 110% (an area where growth is desirable).
Maybe, someday, we will get to the point where we can point to "good" reductions in spending for children. Until we are able to produce these more complete pictures, the fair-share analysis will continue to be an important tool.
The high cost of remediation: Cost per child served
One fairly simple analytic tool that helps make the case for investments in prevention is the (dramatic) difference in spending per child in remedial programs vs. "prevention" programs. An excellent example of this analysis appeared in the New York City Gap Project FY 1992, and served to make the case for "pay me now or pay me later" investments in children. The last chart in this publication shows "Average Expenditure per Youth," with the Department of Corrections exceeding $35,000, and the Parks Department coming in under $3,000. A similar analysis is presented in the San Francisco report Follow the Money. (Both analyses appear in Appendix F.)
This kind of analysis anticipates the cost of bad-results approach (which is possible in later developmental stages), and makes the simple point that broad-based supports for families and children are cheap compared to the costs when things go wrong.
Intended and unintended shifts in spending priorities
With two or more points in time, it is possible to show shifts in spending priorities by department, and therefore by implication, by type of service. The Child and Family Policy Center conducted such an analysis for Iowa expenditures between FY 1983 and FY 1992, showing a significant shift in de facto budget priorities, away from education and human services, and toward corrections and property-tax relief. (See the Iowa entry in Appendix F.) Similar analysis in other state budgets show the well-known increase in the proportion of spending on corrections and Medicaid in the late 1980's and early 1990's at the expense of elementary, secondary, and higher education.
B. A Menagerie of Analyses for Stage II Budgets
Funding by function
The defining characteristic of Stage II budgets is the ability to show spending by function across agency, governmental, and sector lines. This means that we can see, for example, total spending in a state or locality for defined types of services such as child care, education, or juvenile justice; or for defined classes of services and supports, such as remediation and prevention.
The ability to look at spending by function allows a very different kind of discussion about spending to take place: How could we make better sense of spending within a functional category? For example, can we make better sense of the funding and service system for child care, job development and training, and juvenile justice? Can we make better sense of prevention/remediation functions that span service categories? These are complex discussions, and no single analytic view will do more than start the process. But without functional spending information across sectors, progress is confounded by missing basic information about level of effort by whom for what.
Functional summaries also play a somewhat simpler role. They illuminate the financial stakes of different players in specific functional areas. When people realize that they are paying for similar services for similar populations and purposes, they face new evidence that "we are in this together," and perhaps they have a new incentive for working together.
Such analyses also provide an opportunity to face up to the possible duplication of services which our current fragmented system may allow to go unnoticed. When more than one agency is providing the same service, it may not mean that there is too much money in the system. But it may well mean that there is a better way to configure the service delivery system. Functional summaries of spending can help identify where fragmented services can and should move toward more coordinated and integrated service systems.
Revenue maximization
Functional summaries also support another important kind of analysis: revenue maximization. The idea of revenue maximization is simple: find ways to maximize non-general fund revenue, so that general funds can be freed up. Revenue maximization efforts draw on the fact that a given service can sometimes be financed in many different ways. For instance, when financing can be shifted to one of the few remaining open-ended federal funding sources (including Title XIX Medicaid in states that have not adopted quasi-caps under waivers or managed care provisions, or Title IV-E federal foster care and adoption assistance), this can in turn free up general funds for reinvestment into improved or expanded services for families and children. Revenue maximization efforts should not be attempted without a strong commitment to reinvest, since otherwise the freed-up funds will be used for purposes unrelated to children and families.
The reason why functional summaries help with revenue maximization is that they provide a picture of the different ways the same or similar services are being funded. This sometimes points to a way to refinance the services so that they make greater use of open-ended funding sources. This is complex, technical work, which goes far beyond what would typically be included in a family and children's budget. But if a strong reinvestment commitment can be secured, then refinancing candidates that show up in the family and children’s budget may justify further exploration.
Fair share of the load: funding by contributor/payor
As Stage II budgets begin to include complete expenditures from more than one level of government and from public and private sectors, it is possible to begin to show the relative contribution of each contributor to spending for children and families. The Stage I analysis described above showed this picture for the funds that happen to flow through one level of government only. This is far from a complete picture. State budgets usually show only those federal funds that are allocated to or administered by the state. County and city budgets usually show only those funds from the federal and state governments that flow through their budget processes. Neither show much, if any, of the private sector’s funding of family and children services and supports.
Stage II budgets can begin to show the contributions of many different players, the proportion of funding from different sources, and the changes in these proportions over time. This in turn can then be used to support analysis of cost shifting between funders and between sectors.
Cost shifting
When family and children's budgets get to the point that they have funding from all sources (federal, state, local, private) and they have multi-year information, then we can begin to look at the relative share across sectors, and we can look at trends in cost shifts between sectors. This information has potential utility in the advocacy process and in the process of jockeying for the best budget position in intergovernmental relations. It can help make clear who is pulling their weight, and where weight is being left unpulled. When information about fund source is combined with information by function, it is possible to look at cost shifting by function. This perspective will become increasingly important as the federal block grant structure enters the next recession, and federal funding begins to decline as a percentage of total spending for welfare and other block-grant-supported programs.
Need and gap analyses
Since Stage II budgets provide pictures of spending by function across departmental and across funding boundaries, they allow us to begin to see the nature of our total investment in certain types of critical services and supports for families and children.
Currently, for example, it is extraordinarily difficult to see total spending for child care at the state or local level. If a family and children's budget can come close to summing child care spending across the community, then we can develop a more credible assessment of how available resources stack up against need. Similar kinds of analyses may become possible for affordable housing, job training, health and mental health services, etc. Currently, expenditures for these functions are spread across many different agencies and budgets, and making sense of the total use of resources, and its relation to total need, is difficult.
Adequacy in relation to other jurisdictions
Functional summaries open up the possibility of new kinds of cross-jurisdictional comparisons. Since different states and counties organize and fund services differently, such comparisons are often a matter of apples and oranges. But as more jurisdictions begin to summarize expenditures at the total functional level, we may improve our ability to compare levels of investment.
This is more than an academic matter. As businesses consider where to locate, the degrees of local investment in education, workforce development, and child care are becoming more important considerations. Jurisdictions with advanced family and children's budgets may actually gain a competitive edge through their ability to demonstrate favorable comparisons, or commitments to remedy under-investment.
Per capita analyses are particularly useful for assessing the relative adequacy of funding for family and children's service functions. Such analyses can compare one state to another, local jurisdictions to other jurisdictions within a state or in other states, or local jurisdictions to state averages. Per capita analyses are usually done in terms of total population, since such data are more current and easier to obtain. But, where data are available, analysis on a per-child basis may be more compelling.
It is important to note that such comparisons may need to be adjusted to reflect differences in cost of living and wage rates, which can account for significant differences in the relative cost of service between jurisdictions. It is possible to use state or city consumer price index or income data to make such adjustments. Per capita analysis (without cost of living adjustments) is illustrated in the Illinois budget: Dollars and Sense. (See the Illinois entry in Appendix F.)
Candidates for fund pool consolidation
Another important by-product of a Stage II family and children's budget is the possible analysis of candidates for fund consolidation. Considerable work is under way across the country on the matter of "devolution" of funding and decision-making from state to local levels. Much of this work is hampered by a "throw it all in the pot and hope for the best" approach to the creating of fund pools to promote flexibility. A different approach involves the systematic consideration of "natural clusters" of funding which are good candidates for fund pools. A natural cluster is a package of funding that brings with it a natural set of incentives to do better. There are two kinds of natural clusters: prevention/remediation clusters and functional clusters. With prevention/remediation clusters, the natural incentive is to save on remediation so that there is more to spend on prevention (or so that total expenditures may be reduced). The most common prevention/remediation cluster in the current service system is found in health managed care. We have come to understand some of the power of this incentive in the ways that managed care has worked, or not worked, to promote money for prevention and profit in the health field. Other prevention/remediation clusters have been used to change incentives in child welfare, mental health, and juvenile justice. With functional clusters, the natural incentive is to provide service more efficiently, so that we can provide more of it. Stage II budgets allow both functional and prevention/remediation clusters to be presented and used as a basis for consideration of devolution.
C. A Menagerie of Analyses for Stage III Budgets
Cost of bad results
The single most important analysis made possible by Stage III budgets is the "Cost of Bad Results Analysis." This includes analysis of the total cost of bad results as well as such costs by payer.
The idea of costing "bad" results starts with idea of "good" results. As discussed in Section II, results are conditions of well-being we hope to achieve for children, families, and communities. They are such things as children born healthy, children ready for school, children succeeding in school, and children staying out of trouble.
Bad results are the opposite, that is, results we hope to avoid. Much, if not most, government spending for children and families, other than elementary and secondary education, is to remedy bad results: children born unhealthy, children not ready for school, not succeeding in school, not staying out of trouble. The costs of these unwanted results show up in both governmental and non-governmental expenditures. It is possible to measure and track these expenditures, and to begin to frame our social and fiscal policies in terms of reducing the growth of these costs.
One approach to doing this work involves answering the questions, "What costs exist today because we are not getting the results we want?" or "What costs would go away if we got 100% good results?" This does not mean that we can ever eliminate the costs of bad results. The matter at hand is progress, not utopia. In this case, the definition of progress will be our ability to reduce or "turn the curve" on the cost of bad results.
This, in turn, leads to a second question: "What expenditures are embedded in the total cost of bad results that are now devoted to turning the bad results cost curve?" This starts to get at the portions of our agenda for children and families that could be part of the solution and not the problem.
By itself, an analysis of the cost of bad results may not mean or do much. As part of a larger effort to reshape budgeting and decision-making systems to focus on results, it could be a powerful new tool in a more disciplined approach to deciding about investments in families and children. The cost of bad results shows what the financial stakes really are—no punches pulled: What are we paying each year? Is this an acceptable amount? What can we, and what should we do about it? It sets up the most important part of children's budgeting: consideration of what it means to invest in real solutions at scale.
There are only a few examples of cost of bad results type analyses in state and local family and children's budgets. These are based mostly on a prevention/remediation categorization of spending. Contra Costa County divides county programs into three categories: Prevention/Early Intervention Services, Crisis/Safety Net Services, and Remediation/Self-Sufficiency Services. The latter two categories make up nearly 95% of the total spending included in the family and children’s budget, and are presented as costs incurred "when positive outcomes are not achieved." A similar analysis can be found in the children's budget in Los Angeles, which includes "prevention" as one of eight functional categories, and in the analysis in the Iowa Kids Count Quarterly Summer 1994, which shows "prevention" spending as 2.8% of total non-education spending. These analyses are the precursors of the multi-year cost of bad results trend analysis discussed above.
Cost of bad results analysis provides a chance to avoid an important trap associated with classifying programs as prevention programs or remediation programs. The rhetorical commitment to prevention is so widespread that every manager of every service likes to think of their program as contributing to prevention in some way. Even "deep end" services, like prisons, have program components devoted to rehabilitation, education, and job preparation which can be viewed as preventing recidivism, and therefore, crime.
So when we pose questions in terms of "prevention and non-prevention," every program claims that it should be counted as a prevention program. The discussion can quickly descend into a useless debate about funding for "good" programs vs. "bad" programs. Having programs compete to be designated as prevention-oriented completely misses the point of the cost of bad results analysis.
The cost of bad results analysis is intended to identify the costs associated with bad results which we wish to reduce. It is possible to think of this analysis in two stages. First ask and answer the question: "What expenditures exist today because we are not getting the results we want?" When the question is asked this way, then the costs we must identify include whole programs which exist because children are not healthy or not succeeding in school; or because families are not stable or self-sufficient. The TANF program, for example, exists in its entirety because all families are not self-sufficient. Another way to think about this first question is to consider what expenditures would disappear entirely if we achieved all good results. This total set of expenditures represents the cost curve we wish to turn.
The second question to be asked is: "What expenditures, embedded in this total, are now devoted to turning the cost curve?" This is the point at which we consider the employment and training components of TANF devoted to reducing the long-term costs of dependency, or the immunization program within Medicaid devoted to reducing long-term costs of remediating health problems, etc.
We have, in essence, asked the prevention/non-prevention question in a way which does not stigmatize programs or create false incentives to categorize expenditures one way or another. It moves us beyond the potential trap of labe