Policy Analysis

Parliamentary Development Project For Ukraine

12.1997

An Introduction to Policy Analysis /Seminar/ Basic Policy Analysis Trevor Brown Parliamentary Development Project Note: Much of the material in this material is a summary of Patton and Sawicki (1992), Basic Approaches to Policy Analysis and Planning Introduction to Policy Analysis Imagine that you are a legislator and that you are presented with the following Proposal to Raise Alcoholic Beverage Excise Tax Rate

proposal from the administration.

PROBLEM: While some studies have demonstrated that moderate consumption of alcohol may actually improve health, the majority of medical literature reports that excessive use contributes to brain damage, cirrhosis of the liver, birth defects, heart disease, cancer of the liver, and a number of other adverse health conditions. These alcohol-related conditions play at least some role in over 100,000 deaths per year. Yet the most dramatic consequence of alcohol abuse is the large number of highway fatalities caused by drivers under the influence of alcohol. In 1997 over 23,000 of the approximately 53,000 highway fatalities were caused by drivers who had been using alcohol, and 15,000 of these deaths were innocent bystanders. GOAL: Reduce alcohol related fatalities. ADMINISTRATION SOLUTION: Recognizing these adverse external effects of alcohol consumption, and seeing a potential for generating substantial public revenue, the administration is proposing an increase in the excise tax on alcoholic beverages from 10% to 20%. Given what has been presented in this proposal, and assuming that the legislators’ own knowledge of the affects of alcohol consumption extend only so far as his or her personal experience, does the legislator have sufficient information to vote for or against this proposal? For most legislators this is not enough information to make a well-informed types of information would the legislator need?

decision about the proposal. What other

1. An assessment of the veracity of the administration’s definition of the problem. For instance, is the evidence from medical studies sufficient to demonstrate the linkage between alcohol consumption and the fatalities indicated by the administration? 2. An assessment of the administration’s goal. Is saving lives the only goal at stake? What about people’s liberty to drink what they please without the burden of government taxation for consumption? What about the right of right of producers to manufacture and sell alcohol? 3. An assessment of the administration’s solution. Will raising the price of alcohol through taxation reduce the consumption of alcohol? Will the reduction in alcohol consumption be sufficient to reduce the number of fatalities? In addition to these basic questions, there are other questions, which are

important to consider as well.

4. What other solutions are plausible? 5. Who is affected by this solution? 6. Is it possible to implement the proposed solution? 7. Is it possible to pass such a solution through the legislature? 8. What are the political impacts of this solution across the broader electorate? 9. Is this solution legal and constitutional? http://iupdp.org

Powered by Joomla!

Generated: 28 October, 2007, 15:57

Parliamentary Development Project For Ukraine

In most democratic legislatures, deputies require this kind of information in order to make a decision about whether to support a policy or not. However, in most cases legislators have little time to do the research necessary to answer these questions on their own. As a result, they often rely on others -- policy analysts -- to provide them with the information they need to make decisions. The product of policy analysis then is advice to a client -- in your case a legislator -- that relates to public decisions. The product of policy analysis is not a single answer or a definitive solution. Instead, policy analysts provide clients with the range of policy options available, and provide an evaluation of these options based on the criteria of choice supplied by the client. The client then uses this information to inform his or her decision. The basic problem is that legislators are often faced with decisions about highly complex problems, but unfortunately the time they give analysts to provide them with the information they need is typically very short. Consequently, policy analysts have to adopt relatively quick and efficient techniques to analyze problems and solutions. Furthermore, while basic research analysis, like that conducted by an academic institution, is characterized by the routine steps of exploration and accepted standards of the scientific method, the most compelling feature of basic policy analysis in the legislative setting is whether the consumer understands it, is able to follow its logic and as a result is able to formulate better policy. Often this precludes highly complex analysis and presentation. Types of Policy Analysis There are essentially two basic types of policy analysis, each of which I. Descriptive: historical analysis of past policies or the evaluation

has two basic subcategories

of a new policy as it is implemented

A. Retrospective: analysis referring to the description and interpretation B. Evaluative: analysis referring to program evaluation (were the

of past policies (what happened?)

purposes of the policy met?)

II. Prospective: analysis that focuses upon the possible outcomes of

proposed policies

A. Predictive: analysis that makes projections about future outcomes

resulting from adopting particular alternatives

B. Prescriptive: analysis which recommends actions because they

will bring about a particular result

The techniques which we will explore in this training can be used for any of

these types of basic analysis.

There are a variety of approaches to policy analysis. The basic theoretical approach is called the rationalist model in which problem definition leads to the identification and evaluations of policy alternatives followed by policy implementation. When time, information and resources are available analysis often takes this form. What this seminar will do is take you through some of these basic steps, but show you strategies you can employ when there is not time available for long instensive researched analysis. Decision making in parliaments is often constrained by time, and as noted above, analysts are often required to give advice to policy makers in incredibly short periods of time. Basic rationalist approach In the course of this training we will briefly explore stages 1 - 4. In addition we will discuss presenting your results in a manner that is easily understandable for your client. Your basic goal as an analyst is to provide your client with the information he or she needs to make an informed judgement and therefore doesn’t get caught in a major error and can contribute to a more enlightened debate about particular public policies. At the end of this seminar we hope that you will: 1. Learn to recognize sitatutions in which specific basic methods can be applied

quickly and appropriately.

2. Become aware of using methods of analysis and designing approaches to policy 3. Learn how to communicate the results of analysis to appropriate decision

makers.

Before we move on to examine the first four stages, it is important to consider Patton and Sawicki (1992), two experts in the field of policy analysis. 1. Learn to Focus Quickly on the Central Decision Criterion (or Criteria) of

problems.

some practical principles suggested by

the problem.

2. Avoid the Tool-box Approach to Analyzing Policy 3. Learn to deal with uncertainty http://iupdp.org

Powered by Joomla!

Generated: 28 October, 2007, 15:57

Parliamentary Development Project For Ukraine

4. Say it with numbers 5. Make the Analysis simple and transparent 6. Check the facts 7. Learn to Advocate the Positions of others 8. Give the client analysis, not decisions 9. Be aware that there is no such thing as an absolutely correct, rational,

and complete analysis

10. Always remember that in a parliament, politics matters. Six Basic Steps for Policy Analysis Step One: Verify, Define, and Detail the Problem Policy analysis is fundamentally a process of finding a range of solutions to problems. Consequently the first step is to understand the problem in question, verify that it exists, define it, and detail its specific components. Problem definition is often difficult though because the objectives of the client may not always be clear or may be in conflict. Furthermore, as conditions change, the problem itself may change. As a result, the policy analyst must continually ask whether the problem that precipitated the analysis still exists. Furthermore, as the figure above shows, at various stages of the policy analysis process the analyst must return to the problem definition stage to ensure that the path they have chosen to solve the problem is appropriate and that the definition of the problem has not changed. Before formally undertaking your analysis of the problem here are some questions

you need to consider as analyst:

1. Does the problem exist? Think again about the example presented in the introduction. The information presented in the administration’s description of the problem indicated that “alcohol-related conditions play at least some role in over 100,000 deaths per year.” In a country like the United States with a population around 250 million people, 100,000 deaths represents approximately 0.0004% of the population. Is that a sufficient number of deaths each year to consider this a problem deserving of the legislature’s attention? We need more information. In particular we need to know what percentage of people who die each year die from alcohol related causes. In 1996, 3.3% of deaths were as a result of alcohol related causes. Is this enough information? It might also be helpful to know how alcohol related deaths stack up against over causes of death. See the table below. Deaths as a percentage of all causes in 1997

Cause of Death % of All Causes Circulatory Diseases

56.5

Malignant Cancers Murder

13.6 12

Alcohol Related

3.3

Auto Accidents HIV-AIDS

1.5 0.05

As a basic rule of thumb here are some factors to consider in deciding whether

the problem is in fact a problem.

1. individual vs. societal Is the problem concentrated on a small group of people, or does it have broad impacts across society? For instance, are alcohol related deaths concentrated only on those who consume alcohol, or are there spill over affects onto other people? 2. serious vs. benign Is the problem serious or benign? Are the impacts of the problem sufficient http://iupdp.org

Powered by Joomla!

to warrant the attention of the legislature no Generated: 28 October, 2007, 15:57

Parliamentary Development Project For Ukraine

matter how broad their impact? 2. What are the different perspectives on the problem? People have different views about what constitutes a problem. These differences can often be attributed to differences in values. Values, or general beliefs about the relative worth of items or behaviors, are at the heart of problem definition. Problems can be verified, defined and detailed only in relation to the values of the groups and individuals involved. Identifying the values of the groups and individuals involved leads to the goals and objectives in solving the problem. Consequently as an analyst you need to understand the values that underlie problem definition and where they came from. Unfortunately, values are extremely difficult to measure. Nevertheless you need to make an attempt. The values underlying the problem will have a primary impact on the range of policy alternatives available to address the problem. For instance, return to the alcohol problem presented in the introduction and imagine that the problem was presented in less pragmatic terms and instead in moral terms. PROBLEM: The consumption of alcohol is a moral sin. It corrupts families and breeds distrust among husbands and wives. In our cities alcohol plays a major role in the deterioration of parks and public spaces since empty bottles and drunkards populate these areas. It is well known that excessive alcohol consumption often leads to death. GOAL: What values underlie the two different conceptions of the problem? Will the goal be different in this second example? What impact will this have on the range of alternatives that can be considered in addressing the problem? 3. What are the components of the problem? Before considering solutions to a problem you need to make an initial cut at thinking about what the components of the problem are. What factors are we most concerned about? Does the problem have a singular negative impact or multiple impacts? Are there multiple causes of the problem? What behaviors are involved? What populations are impacted? For instance in the alcohol example the components of the problem include: Impacts Primary Causes Secondary Causes Population Impacted Adverse Health Impacts Brain Damage Alcohol Consumption Purchasing and Production Primary Consumers Cirrhosis Alcohol Consumption Purchasing and Production Primary Consumers Birth Defects Alcohol Consumption Purchasing and Production Primary Consumers Heart Disease Alcohol Consumption Purchasing and Production Primary Consumers Liver Cancer Alcohol Consumption Purchasing and Production Primary Consumers Death From health impacts Alcohol Consumption Purchasing and Production Primary Consumers From motor accidents Alcohol Consumption and Operating a motor vehicle Purchasing and Production Primary Consumers and Innocent Bystanders 4. Can the problem be addressed by policies over which the relevant decision

maker (often the client) has control?

The next question to ask is whether or not the decision maker in question can do something about the problem. In order to answer this question you need to think about the factors which lead to the negative impacts. The decision maker may have more influence over some factors than others. Next you need to think of the tools available to the decision maker and what impact they will have on causes of the problem. For instance, in the alcohol example, the administration has proposed increasing the tax on alcohol. Which of the factors listed in the chart does this impact? What behavioral effect is the administration hoping for? What assumptions is the administration making? 5. Is enough information available to conduct an analysis? The final step to ask yourself before proceeding is whether or not you have sufficient information to conduct the analysis. It is a good idea to make a list of all the questions you will need to answer in performing the analysis and then identifying what type of information you will need to answer these questions. Numerical data is typically easier for the client to understand so it is important to find out if needed quantitative data exists and is in a usable form. Steps in Developing the Problem Statement After you answer these introductory questions, the next step is to develop at this stage of policy analysis is to: http://iupdp.org

Powered by Joomla!

a problem statement. The central challenge Generated: 28 October, 2007, 15:57

Parliamentary Development Project For Ukraine

-

state the problem meaningfully to eliminate irrelevant material to say it with numbers, anecdotes and scenarios to focus on the central, critical factors to define the problem in a way that eliminates ambiguity

The ultimate goal is to frame the problem in concrete terms and develop a statement understanding of the problem’s political and technical dimensions.

that gives the client a firm

1. Think about the problem This is the initial cut based on the sources of information at hand at defining the problem and identifying the values at stake. Here you are essentially answering the questions you have answered above. 2. Delineate the boundaries of the problem Specify the problem’s location, the length of time that it has existed, and historical events that have shaped the problem. We must be aware of the connection of the problem under analysis to other problems. As these other problems are resolved or as they worsen, our analysis can be affected. 3. Develop a fact base Problem definition requires some basic information. Consult multiple sources of information and use several estimating techniques. Verify information and compare them with other established facts and benchmarks. The facts to collect can be derived from the problem statement. 4. List goals and objectives Acceptability of possible solutions will depend on the goals and objectives of the respective actors. Some goals and objectives may have to be stated tentatively and revised as the analysis progresses. Others will have to be deduced. Nonetheless, a general goals statement and list of objectives must be prepared, or we run the risk that the problem will be ill-defined. The objectives must be stated so they can be measured, and the measures for each objective must be specified. 5. Identify the policy envelope The policy envelope is the range of variables considered in a problem, and these will affect the alternatives eventually examined. Sometimes the policy envelope is prescribed by the client, sometimes it is determined by the setting in which the analyst is working, and other times it is defined by time and resources available. The analyst will also receive important clues about the size of the policy envelope from the community, attentive groups, and other actors. The analyst must locate the leverage points where policy can be affected and the relevant decision makers. If these policysensitive variables are not included in the envelope, the superior policy may not be identified, and if identified, may not be implementable. 6. Display potential costs and benefits Report in narrative, chart or tabular form the potential costs and benefits of the problem to the actors and interested parties. Indicate what each actor will gain or lose if the problem is resolved. Here the attention is not on the impact of alternative solutions, but on the range of views about what the problem is and what a theoretical solution would be. 7. Create Operational definitions One of the principles of analysis is to “Say it with numbers.” That is where ever possible, measure the problem being analyzed or the alternative solutions being assessed so that a portion of the analysis can be done quantitatively. This process of expressing a problem statement or objective in measurable terms is referred to as creating an operational definition. In every case the analyst should avoid ambiguity, use multiple measures to define a complex concept, use qualitative as well as quantitative measures, employ primary data whenever possible, and be cautious when using secondary data. 8. Review the problem statement Has the problem been stated in a way that will allow for action? Have enough insights been developed to give clues about possible alternatives? Challenge the assumptions that you have made. The next page presents an example of a chart which summarizes the type of information http://iupdp.org

Powered by Joomla!

that goes into creating a Generated: 28 October, 2007, 15:57

Parliamentary Development Project For Ukraine

problem statement. Such a chart might be useful for

your own policy analyses.

Problem Analysis Chart

Brief Statement of Problem: Medical studies have demonstrated that the consumption of alcohol leads to numerous adverse health affects. 100,000 deaths each year are attributed to alcohol-related causes. Problem Boundaries Location - Nationwide Length of Time Existed - Studies date back to 1950s Other Relevant Problems - Job-loss - Addiction Fact Base - 100,000 people die each year from alcohol related causes - 23,000 of 53,000 auto deaths can be attributed to driving under the - 15,000 of these 23,000 are innocent by-standers - Alcohol related deaths account for 3.3% of all deaths in 1996

influence

Goals and Objectives GOAL: Reduce alcohol related fatalities. Objective 1: Reduce the consumption of alcohol. Objective 2: Reduce the purchase of alcohol. Policy Envelope Factor 1 Quantity of alcohol consumed per capita Factor 2 Cost of alcohol Factor 3 Where purchased Factor 4 Affects of different types of alcohol Potential Costs - Individual freedom to consume alcohol curtailed - Reduced sales for manufacturers of alcohol components - Reduced sales for producers, bottlers, and distributors of alcohol - Potential job losses Potential Benefits - Lives saved - Increased life expectancy - Improved health - Lower health care costs - Safer roads Operational Definitions http://iupdp.org

Powered by Joomla!

Generated: 28 October, 2007, 15:57

Parliamentary Development Project For Ukraine

Annual Alcohol Consumption: # of liters of alcohol consumed in a 12 month population over the age of 18.

period divided by the portion of the

Back of the Envelope Analysis Quantitative analysis will be part of some problem descriptions. For others, four basic methods for determining unknown numbers:

numbers will have to be derived. There are

1. Look up the number is a reference source 2. Collect the number through a systematic survey or other investigation 3. Guess the number 4. Get experts to help you guess the number Once these base data have been found, develop back-of-the-envelope calculations using ordinary mathematical operations -- usually adding, subtracting, dividing, or multiplying. Back of the envelop calculations can help answer basic policy questions such as the following: 1. How many persons or families are affected by this problem? 2. How much does a service cost per unit delivered? 3. How many clients can be served with a given budget? 4. By the year 2000, given existing trends, how many persons will be eligible

for a given program?

The back-of-the-envelope calculation method amounts to establishing some of the key dimensions of the problem and checking numerical estimates against known reference points. Notes on Creating Operational Definitions The process of expressing a problem statement or objective in measurable terms is referred to as creating an operational definition. Serious problems can arise if the step that turns ideas into measures is not done with care. There are three basic concerns in creating operational definitions: 1. Reliability: will the measure produce unambiguous data? 2. Fractional measurement:: does the measure cover only part of the meaning 3. Ecological fallacies: are the conclusions generalized to measurement units

of the concept or all of it? other than the one being studied?

A valid operational definition will: 1. State the concept in unambiguous terms. 2. Give attention to qualitative as well as quantitative measures. 3. Take account of spillovers 4. Use primary data when possible. 5. Use data collected for the unit of analysis under study. 6. Draw only conclusions warranted by the unit of analysis. 7. Avoid speculating about individual characteristics from group data. 8. Develop policy aimed at real social and political units. Step Two: Create Evaluative Criteria How will an analyst know when the problem is solved or when an appropriate or acceptable policy is identified? How will possible politics be compared? Any proposed policy will have a variety of impacts and may affect various groups differently. Not only will policies that are acceptable to one group possibly be unacceptable or harmful to another group, but a policy that appears acceptable when judged on the basis of cost for example, may become unacceptable when http://iupdp.org

Powered by Joomla!

Generated: 28 October, 2007, 15:57

Parliamentary Development Project For Ukraine

its environmental impacts are considered. In order to compare, measure, and select among alternatives, relevant evaluation criteria must be established. Criteria are what we use to guide decision making by helping us to compare alternatives. We began by defining the problem and determining the general goals and more specific, measurable objectives of our client. Once we know what is desired, we seek alternative ways of attaining the goals and objectives. The consequences of selecting each of the various alternatives are then estimated. Some commonly used measures include: - cost - net benefits - effectiveness - efficiency - equity - administrative ease - legality - political acceptability Where does the analyst obtain the decision criteria? Sometimes the client provides them, either directly as measures, or indirectly through a statement of goals or objectives. In the former case it is the analyst’s job to make the criteria specific enough so their attainment can be measured. In the latter case you have to deduce criteria and confirm them with the client. The analyst will have to identify those criteria that are central to the problem under analysis and most relevant to the key participants in the decision process. The analyst seeks criteria that meet these specifications, but sometimes criteria are determined by the data that are available. Nonetheless, specifying evaluative criteria and deciding the dimensions along which the alternatives will be measured cause the analyst to clarify the values, goals, and objectives of the interested and affected parties and to make explicit both the desirable and undesirable outcomes. By stating the criteria in advance, we are setting up rules that have to be followed when comparing alternatives. Stating evaluation criteria early also helps avoids the temptation to rationalize preferred options later. New criteria may be unearthed during later stages of the analysis, but the analyst must explicitly recognize the addition of those criteria. Goals, Objectives, Criteria and Measures Goals: Formally and broadly worded statements about what we desire to Objectives: More focused and concretely worded statements about end a client population specified.

achieve in the long run. states, most usually with a time dimension and

Criteria: Specific statements about the dimensions of the objectives that will be used to evaluate alternative policies or programs. Important dimensions include: cost, benefit, effectiveness, risk, political viability, administrative, ease, legality, uncertainty, equity and timing. Measures: Tangible, if not quantitative, operational definitions of criteria. Each criterion should have multiple measures associated with it. Measures can be used comparatively over time on the same problem, or over space, comparing different places with the same problem in order to judge how close alternatives come to satisfying stated criteria. As an example, we will use the problem of water pollution in the rivers of a metropolitan area. The goal, often stated by public officials, is to “clean up our rivers and make them safe and usable.” One objective could be to make the segment of the Challooche River from the county line north to Beaver Dam fit for fishing. One group of criteria addressing this objective would relate to effectiveness: would the program actually make it possible to fish for specific kinds of fish? Changes to water quality, the amount and variation of flow, and other uses of the water and stream banks might be relevant. One measure of effectiveness in altering water quality might be the level of dissolved oxygen present in this particular river segment. Establishing Evaluative Criteria Clearly it is not sufficient for the analyst to devise a list of criteria and determine their relative importance. These activities must be undertaken in concert with groups and individuals involved in the decision process. The first step is to define and establish acceptable and useful criteria. The role of the analyst includes helping the decision maker to clarify what is sought and to define the objectives so that alternatives can be designed. If the objectives has not been specified properly, the alternatives designed will be inappropriate and the recommended alternative will not solve the real problem. In practice, however, decision makers may not be willing to establish clear evaluation criteria. They may not wish to define such terms as equity or efficiency, and they may not want the analyst to predict how much unemployment would be reduced by a favored program or the relationship between the benefits from two competing programs. The reasons are obvious. Politicians must serve the interests of diverse groups; they do not want to support the program http://iupdp.org

Powered by Joomla!

Generated: 28 October, 2007, 15:57

Parliamentary Development Project For Ukraine

favored by one group at the risk of alienating another group. they know from experience that programs are rarely implemented as designed and that outcomes are seldom as projected, and they do not want to be tied to one solution or program. As times change, and the needs of constituencies change, politicians want to be able to respond without being committed to old ideas. The second common problem analysts have with politicians is their unwillingness to discard multiple, conflicting objectives. Often politicians like to have it both ways in order to please groups with competing values or conflicting agendas. Because decision makers do not wish to establish clear evaluation criteria does not mean that they necessarily want to be free to change directions from week to week. Rather, they want to be able to head in a general direction, to aim toward a goal, but use a variety of ways to get there. You may have to initiate the analysis with only a vague idea of what is sought, clarifying the details of the alternatives as the analysis progresses. Arriving near the target, not necessarily on it, is often acceptable to the decision maker. Analysts and evaluators, on the other hand, would like to have specific criteria against which to measure progress toward goals and objectives. In the end the primary source of program objectives and evaluation is one’s client or employer. However, clients may need help expressing these objectives and criteria, and even where the client has stated explicit criteria, the analyst may feel it is necessary to introduce additional objectives and criteria or revise existing ones. Commonly Employed Evaluative Criteria Technical Feasibility: measure whether policy or program outcomes achieve

their purpose.

Criterion: - effectiveness Subcriterion: - direct impact - indirect impact - long-term impact - short-term impact - adequacy Economic and Financial Possibility: measure, first what the programs

cost, and second, what they produce for benefits.

Criterion: - tangible and monetizable costs - tangible and monetizable benefits - change in net worth - economic efficiency - profitability - cost effectiveness Political Viability: measure policy or program outcomes in terms of makers, legislators, administrators, unions, etc.

impact on relevant power groups such as decision

Criterion: - acceptability - appropriateness - responsiveness - legal criteria - equity Administrative Operability: measure how possible it is to actually implement political, social, and most important, administrative context.

the proposed policy or program within the

Criterion: - authority - institutional commitment - capability - organizational support Political Analysis Political factors need to be considered throughout the policy process, but this may not always be possible, especially for beginning analysts who have little political experience or for certain narrowly defined problems. But whenever http://iupdp.org

Powered by Joomla!

Generated: 28 October, 2007, 15:57

Parliamentary Development Project For Ukraine

possible, political factors should be part of problem definition, criteria selection, and alternatives generation, evaluation, and display. Analysts must look at political issues as an integral part of the process, learn terminology to communicate about these political factors, and use consistent methods to report, display, and analyze political issues. Answering the following questions will help clarify political problems. Actors: Who are the individuals or groups usually concerned about this expected to become involved in the current problem?

type of problem, and who might reasonably be

Motivations: What are the motives, needs, desires, goals, and objectives various actors?

of the actors? What will it take to satisfy the

Beliefs: What does each key actor believe about the problem? What are participants? What do you see as desirable means and ends?

the attitudes and values held by the various

Resources: What does each actor have that can be used to get what is wanted? Resources are often monetary, but can take the form of management skills or a sense of timing. Some individuals and groups are able to better use their resources than others. Which actors are most able and likely to get what they want? Sites: Where will decisions be made? By whom? When? Sites can be identified from legislative intent, administrative procedure, and past conflicts. Sometimes decisions are broken apart and are made at several sites. Step Three: Identify Alternative Policies By this stage in the process, the analyst should have an understanding of the values, goals, and objectives not only of the client but also of other involved parties. Knowing what is sought, and having identified the criteria to be used in judging alternatives, help the analyst generate alternative policies. The analyst will most likely already have a list of possible alternatives. If the client had a favorite policy in mind when assigning the problem, the analyst may have deduced this. The no-action alternative (continuation of the status quo) and minor modifications to the status quo are policies that deserve consideration. During the problem-identification stage similar problems elsewhere may have been located, and the solutions tried in those locales can be considered. The analysts who were involved in those problems may be able to suggest other possible policies. Generating and combining alternatives may reveal aspects of the problem not identified earlier. Examining alternatives used by others in analogous situations may provide additional insights into the problem. it may be necessary to redefine the problem because of this new information, and the reformulated problem statement may lead to a revision or addition of evaluation criteria. The challenge at this step in the analytic process is to avoid settling prematurely on a limited number of options. A common error in generating alternatives is creating so-called solutions that do not squarely address the problem. This may occur because the problem is imperfectly or incompletely defined, the objectives have not been clearly identified, or the analyst has a bias toward a particular solution. It may even result from analytic sloppiness. No matter the reason, if the alternatives or possible solutions that are analyzed do not truly address the problem, then the entire process of analyzing alternatives is a waster exercise. To create feasible alternatives, the problem we are trying to solve must be correctly identified and its various components must be specified. Furthermore, if we are to evaluate alternatives or possible solutions, we must also identify relevant judgment criteria. Although in practice the analyst moves back and forth between evaluating alternatives, designing alternatives, and specifying criteria, we have argued that the policy analysis process should begin with problem definition and the specification of criteria. Having completed these steps, relevant alternatives can be crafted. We believe that this sequence will help the analyst avoid locking in on favored alternatives that do not relate to the problem or that cannot satisfy the evaluation criteria. Sources of Alternatives Where does the analyst obtain alternatives? How are they generated, discovered or created? To begin, the no-action alternative, retaining the status quo, should always be considered as an alternative, and that it is needed as a base-line alternative in cost-benefit analysis, for example. Additional alternatives can be derived from the experience of others with related problems, research findings from these cases, analogy to similar problems, the experience or insights of experts, requirements of authority, the beliefs of participants, legal prescriptions, technical knowledge, and so on. No-action (“Status Quo”) Analysis: In order to make a sound judgment between alternatives we have to invest as much time in analyzing the no-action alternative as in analyzing the action alternatives. Developing a detailed no action alternative and forecasting its results provides a benchmark against which the results of all action alternatives can be measured. http://iupdp.org

Powered by Joomla!

Generated: 28 October, 2007, 15:57

Parliamentary Development Project For Ukraine

Quick Surveys: From a very practical standpoint it is a good idea to let other analysts and friends know you are working on a new problem, for they may have suggestions about alternatives or will remain alert for ideas to pass on to you. Analysts develop a network of friends to call upon, to test ideas on, and with whom to share new ideas. Classmates can form the basis of this network. When faced with a new problem, try these contacts for relevant ideas. Literature Review: It would be a mistake to overlook the literature as a source of alternatives. The literature of the substantive field in which the problem is located (e.g., housing, education, defense) should also be examined as a source of alternatives used elsewhere (both successes and failures) and proposed policies that may not have been tried. Comparison of Real-World Experiences: While searching for alternatives through a quick survey or literature review, analysts uncover both policy ideas and real-world experience. We believe it is important to separate out from this list a special list of real-world alternatives and to compare types of alternatives used in similar settings. The alternative alone may be a valuable idea, but its relative usefulness is enhanced if we can determine why it was adopted, what other alternatives were discarded, whether the alternative was modified after implementation, who supported and opposed it, and how it has fared. A quick first cut at this information can be very useful.

Passive Collection and Classification: To this point we have discussed the search for alternatives as if it were essentially an active pursuit of elusive concepts or new ideas. This is not always the case. Proposals will come from clients or superiors, from advocates of various positions, and from organizations and other interest groups Besides actively seeking alternatives, the analyst should systematically record and classify alternatives suggested by others. Analogy and Metaphor: A possible solution to a problem might be found by examining how analogous problems were solved in the past. Analogy and metaphor have been used in problem solving both to define problems and tot help the problem solvers identify possible solutions or alternatives. Supporters of this method argue that we often fail to find a solution to a problem because we do not recognize that our seemingly new problem is really an old problem. Brainstorming: Brainstorming can be used to conceptualize possible solutions to problems. The modern brainstorm session is a creative conference for producing a checklist of ideas leading to a problem solution. Brainstorming ranges from informal, quick meetings among staff members working on a problem to more structured meetings of staff, experts, and consultants. Pitfalls to Avoid 1. Relying too heavily on past experiences. 2. failing to record ideas and insights as they occur. 3. Locking in on a problem definition too soon. 4. Forming a preference too early. 5. Criticizing ideas as they are offered. 6. Ruling out alternatives through preevaluation. 7. Failing to reconsider dismissed alternatives as conditions change. Steps Four and Five: Evaluating and Presenting Alternatives The linchpin in the policy analysis process is the evaluation of alternative policies and the packaging of policies into strategies and programs. What are the expected impacts of each policy? To what extent does each policy satisfy the evaluation criteria. The step of evaluating alternatives -- comparing them to your decision criteria -- may reveal alternatives that satisfy most or all of the major criteria, and it may reveal others that can be discarded with little additional analysis. Some alternatives will call for further examination. Additional data may have to be collected. During this stage it is important for the analyst to recognize the difference between economically or technically feasible and politically acceptable alternatives. Policy formulation -- the designing and evaluation of alternatives or policy options -is aimed at defining the problem appropriately and finding feasible and effective solutions. The policy evaluation step is also a point in the analysis where we may discover that the problem no longer exists as we defined it or as it was defined for us. Information discovered during the identification and evaluation of policies may reveal new aspects of the problem which may in turn call for additional or different evaluation criteria. Depending upon the analyst-client relationship, the results of the evaluation http://iupdp.org

Powered by Joomla!

might be presented as a list of alternatives, Generated: 28 October, 2007, 15:57

Parliamentary Development Project For Ukraine

an enumeration of criteria, and a report of the degree to which the criteria are met by each alternative. This is not to suggest that the numerical results can or should speak for themselves. Even in this somewhat neutral presentation format, the order of criteria, the sequence of alternatives, and the space given to various options can influence decisions. For a problem that involves quantitative and qualitative information, it is unlikely that the results of the analysis can be presented as a numerical summary. Furthermore, there may be no agreement on the relative importance of criteria. In such a case a comparative matrix could be used to summarize the analysis. Along one axis would be listed the alternatives, along the other the criteria. The cells would contain net benefits and costs. Such a format permits the decision maker or client to assess the options and to use the analysis to make a policy choice. Some clients prefer that the analyst present a strong argument for the superior option, and in some instances the analyst may feel that a particular option has such overwhelming merit that a special case must be made for it. In these cases scenario writing may be the preferred tool, since it allows the analysis to be placed in a larger context, can give life to a dull analysis, can excite, can anger, can move a person to action. Keep in mind the difference between a technically superior alternative and a politically viable one. Sometimes the preferred alternative, in a technical sense, is known, and the task is to deal with political opposition. Use political feasibility analysis to display the pros and cons of alternatives. Rarely will there be only one acceptable or appropriate alternative. Not only will different options appeal to various interested parties, but two or more alternatives may bring roughly similar results. None of the alternatives is likely to be perfect, as problems are rarely solved. More often their severity is reduced, the burden is more evenly distributed, or they are replaced by less severe problems. Policy analysts work under time constraints. As a result, they take shortcuts. They estimate. They overlook alternatives and variables. Most policy analyses are incomplete. Because analysts make recommendations under conditions of uncertainty, these uncertainties must be reported and possible side effects must be identified. The Problem of Multiple Criteria In public problem solving, analysts discover conflict among objectives and among criteria. We discussed earlier the possibility that decision makers might support conflicting objectives. In addition to revaluing the conflicts among objectives and among the criteria used to measure achievement analysis will reveal objectives not considered earlier. As alternatives are examined, the analyst may discover that there is no dominant alternative, that there is no objective agreed upon by all interested parties, and that objectives preferred by affected groups are in conflict. When faced with this problem, analysts typically take one of several approaches to comparing alternatives. They may try to compare alternatives by transforming the costs and benefits to monetizable terms and then evaluating them using this common denominator. They may alternatively seek to defined a so-called high-level objective, a more general one that people can agree upon. this might by derived through voting, or through conflict-resolution techniques. The analyst might get the client to agree on minimum levels of attainment for all but the most important objective, which would then be maximized. The analyst might get the client to rank objectives and then determine which alternative satisfies the top objective. Alternatively, the analyst may decide not to attempt to optimize, but instead select a solution that may not be the best, but one that is good enough and that can be agreed upon. Perhaps the majority of policy solutions, especially quick solutions, are arrived at through this process. In fact, optimizing when there are conflicting or multiple goals often is extremely difficult. In some cases there may be no way to combine or add up the various impacts.

http://iupdp.org

Powered by Joomla!

Generated: 28 October, 2007, 15:57