Essay on Public Policy & Social
Public policies are developed based on the issues that trouble the members of society to the point that there is a necessity to take an action. According to Larry N. Gerston (2010), “issues preceding those policies develop when individuals with similar problems are forced to cope, without solution, for an unacceptable period of time”(p. 22). Public policy problems can be solved by means of effective public policy making. Illegal immigration has become a public policy problem. The problem is becoming more and more acute because of indecisiveness on the part of policy-makers. Thesis statement: Illegal immigration has become a public policy problem, which requires the proper solutions by means of effective public policymaking process, supported by public and political awareness of the significance of the policy for economic, political and cultural development of the nation.
- Outline of the major points
- The scope and nature of the public policy problem
The scope of the public policy problem like immigration or illegal immigration is large and expanding. According to James Hollifield and colleagues, “Illegal immigration also drew attention as a prominent public policy problem”(p.102). The statistical data shows that in 2008, there were about 11.9 million of illegal immigrants in the United States (Illegal Immigration Facts & Statistics, 2014). According to the research conducted by the center of Immigration Studies, if immigration continues to grow at current levels, the population of the United States will increase “from 301 million today to 468 million in 2060 – a 167 million, or 56% increase” (Illegal Immigration Facts & Statistics, 2014). The nature of the public policy problem in concluded in the fact that illegal immigrants have no rights in the United States, but at the same time, they have negative impact on the development of the U.S. economy. Illegal immigration leads to increased unemployment. As a result, current immigration policy is ineffective.
- How the problem came to public and political awareness
The problem of illegal immigration came to public and political awareness because of considerable changes in social, political and economic life of the country. Public awareness of the problem like illegal immigration is supported by certain concerns over national security. Ideological pressures influence public interpretation of current immigration laws (Koven & Götzke, 2010). Americans who oppose immigration policy explain their negative attitude to policy implementation by the threat to unemployment, the threat to cultural identity, and national security. Political awareness of the problem is caused by the functioning of interest groups, such as human rights groups and business groups. As the United States in the nation of immigrants, political parties are aware of the significance of the issue in the political development of the nation (Koven & Götzke, 2010).
- The evolution of the related public policy
The evolution of the related public policy is connected with considerable changes in decision making of policy-makers. The problem of illegal immigration begins as an individual issue, although it is closely connected with “the evolution of personal topics into widely shared aggravation that makes such dilemmas suddenly political and, therefore, candidates for public policy activity” (Gerston, 2010, p. 22).
- Level of government and the actors involved
The level of government involved in public policy making to address the public policy problem of illegal immigration include central and district levels of government. Both central and district levels of government have opportunities for involvement in decision making regarding immigration issues. The actors involved in policy making process concerning illegal immigration include civil society, business community, public and private institutions. These actors have access to the resources that may affect the policy making process.
- The intergovernmental structure and political concerns
The intergovernmental structure plays an important role in policy making process in the United States. Certain political concerns regarding immigration policy should be taken into consideration by policy makers. According to researchers, “concerns about integration into a cohesive national identity have been enhanced by fears about Islamic fundamentalism following September 11, 2001 attacks”(Koven & Götzke, 2010, p.178). As the effectiveness of immigration policy depends on ideologies of immigrants, certain political concerns may be connected with the concept of citizenship or human rights issues.
- Conflicting public opinion and impact on policy solutions
There are conflicting public opinions regarding immigration policy. The majority of American citizens have positive views on immigrants. They consider that “immigrant benefit the U.S. economically and culturally”. However, there are many Americans who state that “they are a net burden on the state” (Albertson & Gadarian, 2013, p. 288). It is known that recent immigration policies were aimed at preventing the spread of clandestine immigration in the United States, as well as guaranteed a “minimum level of rights to immigrants, even to those without a residence permit” (De Montis et al., 2012, p. 1227). Nevertheless, many Americans opposed these policies, explaining the significant role of diversifications in human society, which lead to conflicts between cultures.
- The approaches to policy formulation, adoption, and evaluation
The approaches to policy formulation, adoption, and evaluation play an important role in policy making process. Policy formulation process requires high level of political diplomacy and effective negotiation as this process takes place within a large political arena. Policy approaches to address immigration issues require public involvement. The rational approach is the most appropriate for finding solutions to immigration policy problems as it reflect real-world goals. It is necessity to define the problem, social values and policy goals in order to explore the major policy alternative solutions. Policy adoption process is based on the awareness of the significance of the issue. Policy evaluation should be focused on design, explanation and testing of various methods to address policy evaluation issues, including case study evaluation, needs analysis, realistic analysis and other methods.
- The suggested policy direction (continuation, change or termination) and future impact
The brief overview of immigration public policy problem shows that illegal immigration influences practically every area of public concern. As a result, the suggested policy direction should be based on change of current public policy decisions. It is necessary to make certain changes in order to accelerate the economic development of the nation. Generally speaking, today’s immigration law in the United States requires reformation. According to Briggs, it is necessary to make “labor-market-oriented changes to immigration policy”, which could form a solid foundation for such reformation (qtd. in Peach, 2010, p. 111). Although changes in immigration law cannot have significant impact on the causes of illegal immigration, they may contribute to finding the ways to improve economic development.
Thus, it is necessary to conclude that the public policy problem like illegal immigration is really serious. As public policy problem affect public policy implementation, it is necessary to find the proper solutions to the problem of illegal immigration. The public response to existing problems and the solutions to these problems should be based on positive public experience.
After decades of continuity and incrementalism, significant changes are coming to social policy in Canada.
The foundations of the welfare state were built in the 1960s and 1970s, and Canada has retained many of the major programs and policies put in place during that era. In the decades since, policy-makers have primarily focused on fine-tuning these programs, extending them in some areas and seeking to reduce costs in others.1
Yet many of the assumptions built into the welfare state are increasingly at odds with reality because demographics, technology and society’s expectations for social policy have changed considerably. In this essay I argue that we are on the cusp of a new period of transformative change that will result in what I call the “enabling society.”
The enabling society is not a single idea or a particular form of utopia. It brings together two elements: first, recognition of the major forces that are now reshaping how we think about social policy; and, second, a framework in which social policy can be modernized and adapted to meet the new realities before us. As the term suggests, the purpose of the enabling society is to make social policy more effective in enabling individuals to pursue their goals and achieve their full potential in life.
Reaching for the same high-level goals as the welfare state, the enabling society will seek to harness new technology and new approaches to deliver programs and services that are more individualized, flexible and focused on long-term outcomes over a person’s life. The changes the enabling society will bring about will allow us to finally address problems that have always existed but were beyond the reach of traditional policy instruments in the welfare state. And, importantly, they will do so in a way that provides more efficient program delivery and more effective collaboration between governments and other actors.
The essential feature of the enabling society, once it matures, is a system of “big statistics” that will deliver detailed information to policy-makers, service providers and citizens about the kinds of interventions that will work best for each individual, given his or her particular needs, circumstances or preferences. Although governments have been slow to adapt and transform, major advances in statistical tools and technology over the last several decades now allow us to integrate this evidence-based information throughout the design, delivery and governance of social policy. Just as similar advances have transformed the way companies operate and interact with consumers, increasing the potential for highly sophisticated and predictive microtargeting, the opportunity is great for improving how governments reach citizens, respond to their needs and achieve outcomes for them – and, especially, with them.
Social policy will be able to move beyond addressing the broadly defined needs of broadly defined groups of beneficiaries in a broadly uniform manner, to an approach that is truly individualized and based on a life-course perspective: a view that takes into account a person’s entire life path rather than looking only at their situation at a single point in time. Citizens will benefit from greater choice and better decision-making as they pursue their own goals for learning, health, employment, caregiving and income security.
For governments, this paradigm will place a greater emphasis on working in collaboration, providing transparent data and information, and addressing issues of intergenerational equity. Although this approach will require a new way of doing things, programs will contain self-learning feedback loops so that they continuously improve and evolve over time.
Some readers will understandably be concerned about what all this accumulation of information means for privacy and the spectre of Big Brother. Big statistics is not about increasing the amount of individual data that is collected, nor about using these data to influence behaviour. Rather, it is about using existing data more effectively and in a way that empowers citizens as they make decisions. Privacy and individual freedom will remain paramount values.
From the series of changes described in this essay will emerge an architecture of social policy very different from the welfare state we have come to know. This new model will bring with it the potential of significant benefits: better decision-making by citizens and government; simpler and more efficient service delivery; greater openness in government; and, most importantly, a more effective framework to help individuals fulfill their aspirations.
Some opportunities offered by the enabling society are already here, and many more are within our reach. With some restructuring in the machinery of government, new thinking on the role of government in social policy and a better understanding of the implications for governance and accountability, the path forward is achievable. In this essay I lay out what is driving us toward this new reality, the implications it holds for how programs are designed and structured, and what it will take for governments to complete the transition to an enabling society.
Much of this essay examines the enabling society through the lens of programs that support income security, employment and skill development. Similar conclusions would apply in other areas of social policy, notably health and education.
Getting to the Enabling Society
The conventional wisdom
Countless studies in recent decades have examined the social and economic factors that are shaping social policy. The conventional wisdom drawn from this body of literature comes down to these three observations.
Population aging will put pressure on governments to strengthen long-term care, control the growth of health care costs, support continued labour force growth and reform pensions to make them more sustainable and encourage later retirement.
- Globalization and the shift to the knowledge economy are also central forces, necessitating greater emphasis on policies to enhance competitiveness, innovation and productivity. This, in turn, leads to calls for greater emphasis on policies to increase skill levels and keep workers in the labour force and to remove barriers and disincentives to work.
- More recently, increasing income inequality and the persistence of pockets of poverty and exclusion have drawn greater public attention. People are seeing the growing wealth of those at the top and, at the other end of the income spectrum, at-risk populations who face barriers to economic and social participation that persist over long periods of time. There is, as yet, no consensus on policy responses to address these issues.
- Yet in general, these trends are not seen as reasons to move beyond the current incremental agenda for social policy reform: one focused on restraining the escalating costs of social programs, improving the effectiveness of programs where possible and enhancing programs in only limited ways. But I draw very different conclusions, both about the factors that will drive potential policy reforms and about their effects.
A different understanding of fiscal pressures
The mature welfare state of the 1960s and 1970s created public expectations about the scale of social interventions, and many governments subsequently found the cost of meeting those expectations to be too high. Some critics saw disincentives to work that appeared to be associated with overly generous welfare and entitlement programs. In Canada, much of this concern focused on the design of the unemployment insurance program (later renamed employment insurance, or EI). The recession of the early 1990s underscored the need both for fiscal restraint and for ways to ensure the sustainability of social programs, particularly pensions, for an aging population. By the mid-1990s, it was clear that the retirement of baby boomers two decades years later would leave behind a smaller working-age population to support the pension and social benefits of a larger number of retirees. These demographic changes were projected to have potentially catastrophic consequences for the sustainability of these programs unless changes were made quickly. As a result, over the past 20 years, nearly every OECD country has made major reforms to its pension system.
The maturation of the welfare state resulted in a major increase both in total spending on services and in the share of costs financed by government, especially in education and health care. Since that time, the reforms that were intended to cut costs have not, in fact, resulted in any actual decrease in total spending on these services. Rather, the rate of growth of public spending has been reduced, while costs borne by individuals have grown much faster. Consequently the government’s share in financing these services has gone down while overall spending has gone up.
For example, in post-secondary education (PSE), where Canada’s public share of financing has traditionally been lower than the OECD average, the contribution of governments fell from about 85 percent in 1980 to 69 percent by 2000 (Statistics Canada 2014). In the most recently documented 10-year period, the public share of PSE financing declined further in many countries, including Canada. In 2011, the share of PSE costs in Canada covered by governments stood at 57 percent (OECD 2014). A decline in the public share of costs has been particularly evident in income security programs, with a move toward “individualization” and a greater role for individual savings and contributions, especially in the financing of pension benefits.
In looking to the future, the conventional wisdom still identifies fiscal pressures as a major driver of change in social policy. We often hear the same predictions about the “tsunami” of baby boom retirees that were routine in the 1990s. This language persists even today in Canada (Corbella 2012; Ragan 2011), even though the effects of population aging here are relatively modest and governments brought in proactive reforms in the 1990s to ensure the sustainability of public pension plans.
The tsumami rhetoric is, however, at odds with the evidence. Take health care, where the language of “crisis” is most common. The projected effect of population aging on future health expenditures has consistently been found to be small relative to other cost drivers (CIHI 2013, 2011; Morgan and Cunningham 2011). In labour market policy, we have also been slow to recognize a new reality: in fact, there has been a significant reversal of the declining retirement age over the past two decades, and the employment participation of older workers has increased dramatically (figure 1). Indeed, assuming that the strong demand for older workers continues, there will be enough older people working and earning to offset the potentially negative fiscal, macroeconomic and labour supply effects of population aging until at least 2031 (Hicks 2012).
This does not mean that fiscal considerations are unimportant. Social policy must inevitably play a role in the management of a government’s deficits and debt. However, it does suggest that governments will have much greater flexibility and control over how to adapt social policy to these pressures than is commonly assumed.
Nor does this analysis suggest there will be no crisis in the retirement income system. A crisis is indeed likely within the next decade, but it will not be of the kind many experts predict. As people continue to work longer, often up to and past the eligibility age for most public old age benefits, a serious misallocation of public funds is likely to occur. Since actual retirement ages are rising much faster than the standard age of entitlement for these benefits, there is a potential for huge inequality to occur as individuals who can continue to work do so, while at the same time receiving both age-related pensions (including their old age entitlements through Old Age Security [OAS])2 and earnings that are at or near the peak of their lifetime potential.
The federal government has announced gradual increases in the age of eligibility for OAS (as well as the Guaranteed Income Supplement) from 65 to 67 over the next 15 years, but because of its timing, the changes will not affect most of the baby boom generation and will almost certainly be insufficient to prevent this misallocation of resources. The only way of averting it would be to increase the age of entitlement more and sooner: by five years over the next decade (Hicks 2012).3 Furthermore, even the currently scheduled changes might well be reversed before they are slated to begin in 2023, because both the federal Liberal Party and the federal NDP have pledged to do so.
These realities seem absent from the current debate over pension reform,4 which is almost entirely focused on two other topics: declining pension coverage in the private sector and the sustainability of relatively generous defined benefit pensions in the public sector. These are important issues, but they are not at the core of the retirement income system’s most urgent problem.
To address this issue, changes are needed to take account of the growing diversity in life-course pathways and to provide greater flexibility and choice for individual Canadians.
Greater diversity over the life course
Life-course pathways are the trajectories each person follows through the major episodes and events of life, including schooling, work, forming a family and retirement. Already complex, these pathways will become even more diverse in coming years as individuals and society- grapple with the choices that arise from longer life expectancy and staying in the labour force later in life. Box 1 provides some simple examples of the quite varied arrangements in work, savings, lifestyle and family responsibilities that could follow if current trends toward later retirement and more complex transitions into and out of education and work continue.
Actual life-course pathways are of course more complex than can be illustrated in simple examples.
Similar, if less dramatic, changes in life-course transitions are also likely to occur earlier in life, as work and school come to be seen as more integrated over a person’s lifetime. Such changes will deeply affect many areas of social policy,5 as will be discussed later.
Both the academic world and the policy world have been slow to acknowledge the need for more flexible life-course pathways.
Trends in values, public attitudes and expectations
The shift to the enabling society is not only a response to external pressures. The content of social programs is also being reshaped by changing expectations about what constitutes a better society – expectations that go well beyond the fiscal, social and economic drivers just discussed. How do academics, members of the public and others define “better”?
Perhaps the main message coming from thinkers in the academic community is the need for social policy to adopt a human development perspective, and particularly to help people develop their capabilities over the course of life. There is also a new emphasis on individual well-being and individual choice. This does not conflict with the more collective orientation of policies in the traditional welfare state, but it does alter the perspective from which objectives are set and measured, and it changes how the life course is viewed. The disciplines of welfare economics and gender studies have been particularly influential in this shift in thinking, which is often associated with the work of Amartya Sen (1980). Similar themes have emerged in the theoretical literature on the determinants of human development, in gerontology and in the life-course literature (Marshall 2009; Mayer 2003; McDaniel and Bernard 2011; OECD 2007), as well as in a number of reports from agencies such as the OECD.6
Some academics consider social policy not only through the lenses of skills and human development but also from a wider perspective that looks at the influence of family structures and gender. For example, although more women are working and earning income, the growing tendency for people to marry partners with educational and socio-economic backgrounds similar to their own has reinforced household income inequality (Esping-Andersen 2009), as well as intergenerational inequality. However, greater public investment in early childhood learning could help promote greater equality and socio-economic mobility over the life course. While these life-course perspectives are familiar in the theoretical literature, they have not yet been integrated into mainstream policy thinking.
Important changes are also occurring in the values and attitudes of individuals. Results from various waves of the World Values Survey indicate a clear trend across the developing and industrialized world toward support for values such as tolerance, autonomy and greater participation in decision-making (Inglehart and Welzel 2009; Welzel 2011). These values have become a higher priority for many people as basic needs for physical and economic security have been better provided for in much of the world, and as a result of democratization and the emergence of a knowledge-based economy. Among Western democracies, the reduction of poverty thanks to the welfare state has made an important contribution in supporting this evolution of values.
Canadian public opinion surveys pick up some of these same themes, illustrating an important duality in how we think about priorities and the role of government (figure 2). There is a clear and continued preference for government to take a key role in addressing broadly based societal issues, such as improving access to health care and reducing inequality. Where the public today diverges from the traditional configuration of the welfare state is in the kinds of instruments that are considered the best for achieving the goals of social policy. Today, support is particularly strong for policy tools that actively promote social and economic integration (such as education and training), while it is much weaker for passive income support measures like EI (figure 3).
These trends in values and attitudes have important implications for how we think about each of the four major components of social policy:
- supporting people in times of need
- fostering learning and human capital development
- ensuring support at various stages of life
- addressing inequality
Supporting people in times of need
Meeting immediate needs has always been a core function of social policy. What is particularly new, compared with the thinking that propelled the welfare state, is the greater emphasis on tailoring services to the needs of specific individuals. Modern approaches reflect a greater recognition of human diversity, especially in the extent and severity of barriers and needs that people face. As well, we are learning that current measures of poverty and exclusion may be poor indicators of how needs change over the life course and fail to take into account the resources available to people in their wider families or social networks. For instance, most spells of poverty – even for those in high-risk groups – are relatively short, while a minority of people remain persistently poor for long periods. Measuring poverty on a point-in-time basis does not reflect the very different needs these two groups face.
Fostering learning and human capital development
Traditional social policies have focused on learning skills and on developing human capital from a labour market and economic perspective. Welfare state policies have placed priority on extending access to PSE, early childhood education (in most countries, although less so in English Canada) and adult skills training.
More recent thinking reflects a shift in focus away from the labour market outcomes for specific occupational skills and credentials and toward more emphasis on the learning-to-learn skills that are more appropriate in the information age. Generic skills training for adults used to concentrate exclusively on basic reading and writing skills. Now it is recognized that training needs to encompass a range of transferable skills; in addition to the ability to understand texts of varying levels of complexity, people need financial literacy, computer literacy, numeracy and the capacity to innovate and work in groups. The concept of “skills” now refers to how all individuals acquire, use or lose such skills over life.
To date, this new perspective on skills has mainly been used for research and assessment purposes, as witnessed by the growing use of results of international literacy surveys and the integration of skills into occupational classification systems.7 It has not yet been incorporated into mainstream programming that could help Canadians actually build more up-to-date skills.
Ensuring support at various stages of life
Another goal of social policy is to help people access resources they need at particular times in their lives: student aid, unemployment assistance or old age benefits. In the welfare state, these resources are typically provided as direct income transfers and paid for out of general government revenues. This model redistributes income from all taxpayers to those in need: students, the unemployed and seniors. In recent decades, policy has shifted away from this model and toward approaches that help individuals prefund their own resource needs and that enable life-course pathways that are more flexible.
This individualization trend has shifted the orientation of many programs toward life-course income transfers and savings rather than direct income support. Instead of income redistribution between taxpayers, we see reallocation of resources over the life of each individual, an approach similar to private pensions. Historically, fiscal concerns have pushed policy in this direction, but in the enabling society the impetus will be a desire to increase the control that individuals can exercise over their life choices (Hicks 2013; Marshall 2009; OECD 2007; Sen 1980).
The recent literature identifies five themes that might shape future policy-making (Corak 2013; Esping-Andersen 2009; Gilbert 2005; Heisz 2014; Hicks 2013; OECD 2011).
- A more comprehensive concept of inequality. It is important to broaden our measures of poverty and inequality, which are often based on the distribution of income at a given point in time. These measures need to take account of the important role played by assets, how income changes over an individual’s life course, and income mobility across generations.
- A new focus on equality in family life. Many scholars have extended their analytical framework beyond education and income per se to incorporate how these factors interact with family structure, gender and the related roles and responsibilities within families and households (such as caregiving).
- Recasting equality of opportunity as equality of autonomy. The primary goal of policy is now seen as providing equality of access to the tools that people can use in developing their capabilities in ways that make sense to them.
- New interest in equality of outcomes. Since the emergence of the Occupy Movement and the renewed attention it brought to the impact of higher concentrations of income at the very top, public discourse has turned to a discussion of equality of outcomes, a concept that the welfare state is ill-equipped to address.8
- Intergenerational equity and sustainability. The demographic shift we are experiencing has engendered a new debate about intra- and intergenerational equity and how effectively social policy addresses income and wealth redistribution, poverty and labour market outcomes for the old and for the young.
An intergenerational perspective can enrich social policy in a number of ways. It presents an opportunity to integrate concepts of equality with the values of sustainability and social cohesion. The desire to leave the next generation a world that is at least as good as the one inherited by the present generation (both for society as a whole and for our own children and grandchildren) can be a strong unifying force in society, especially if it is cast in terms of equality of opportunity or autonomy. The pursuit of intergenerational equity could bring us closer to a world that truly values liberté,égalité and fraternité.
A focus on intergenerational issues makes particular sense in a country such as Canada, which attaches such importance to immigration, bilingualism and multiculturalism. One test of the success of bilingualism and multiculturalism is the extent to which generations maintain and pass on their mother tongue and their cultural traditions.
Governance of social policy
Deep changes are also taking place in society’s expectations of the way in which social policy should be conducted and the roles played by the various actors within the system. These governance issues tend to be treated, mistakenly in my view, as a separate topic in the public administration literature, without reference to the close links between the factors that drive the content of policy and those that influence its delivery.
Since the creation of the welfare state, there have been many attempts to improve service delivery, to make government more open and accountable, to involve citizens and organizations in the actual design of programs, and to achieve greater coordination across departments and jurisdictions. More recently, the emphasis has shifted toward seeing the role of government as not only to deliver particular programs effectively, but also to enhance the collective capacity of all society’s actors to achieve better social outcomes.
Third-party providers have long had an increasing role in direct service delivery, with government’s role being increasingly related to coordination, fostering collaboration between stakeholders and outcomes-based management.
Unfortunately, the current base of statistical evidence that can be drawn upon to monitor and enhance collaborative arrangements (horizontal management) is weak. The evidence to support improved performance within a single program (vertical management) is better – but still not strong. Many reforms, once they evolved beyond the pilot or experimental phase, either had limited success or failed entirely for lack of useful evidence. Examples include e-government and open government; shifting policy development from a model of public consultation to one of public engagement/partnership; finding the right balance between decentralized and centralized decision-making within government; joined-up or one-stop-shopping service delivery; incentive pay; and red-tape reduction.9
One of the latest innovations in social policy instruments is social finance, which deploys private capital in outcomes-based funding arrangements to achieve social goals. Even this potentially appealing new tool cannot become a normal way of doing business without enormous improvements in the evidence base of “what works,” and in using this evidence to create incentives for funders (Gold and Mendelsohn 2014).
In the absence of these important elements, many reforms to improve horizontal management have made things worse for vertical management – and vice versa. Of the few successful reforms that can be identified, such as efforts over the past decade to improve citizen satisfaction and citizen-centred service delivery at the federal level (Institute for Citizen Centred Service Delivery 2013; Marson and Heinzman 2009), most10 have been limited to improvements in vertical management. These reforms have not, by and large, been used to develop more effective or collaborative arrangements for service delivery.
Big statistics and “what works” evidence
Despite the profound impacts of the new information and communications technology (ICT) we see all around us, the conventional wisdom still seems to ignore the potential for advances in this field to redefine the interaction between social policy and the individual.
When the welfare state was put in place, the goals of social and employment policy could be met only by using a number of simplifying assumptions based on point-in-time analysis, and taking into account only a few selected individual characteristics. The designers of these programs did not set out to create separate program silos, and they did not intentionally ignore the complexities of people’s lives or deliberately decide to restrict people’s choices about how to develop their capacities. Rather, the design of those programs was the best that was possible given the technology available before the computer age.
The last 50 years have seen huge advances in information processing technologies. However, in the social policy world, these technologies have been used primarily to perform the same tasks, only faster and more efficiently. With few exceptions, the content and structure of social programs, national social statistics and policy governance are fundamentally unchanged from what existed in the 1960s and 1970s.11 The potentially transformative effects for policy of these technological advances are still largely in the future, but we are now at an important turning point.
The current generation of ICT provides the means to transform social policy in a way that will allow it to better respond to the changing values and pressures described earlier. The new technology has already been brought into some applications. But the central feature of the reforms we need is a radically improved base of evidence on “what works,” which is made possible by new big statistics techniques. Big statistics is the next stage beyond the big data techniques that we hear about in the news.
The difference between these two concepts is similar to the distinction between data and evidence. Big data can identify interesting new pieces of information by using powerful tools to sort, mine and examine a multitude of rich but unstructured, disparate data sources. Using similar tools, big statistics adds a conceptual structure to integrate the multiple data sources: big statistics takes what big data techniques have found and inserts this information into standard classification systems, accounting frameworks and other models in order to generate useful statistical estimates.
With big statistics, governments can draw together vast amounts of information about the population, compiled from thousands of administrative files, surveys and censuses, and can present it in an integrated and anonymized form. These data will enable governments to produce detailed estimates of the likely future effects for an individual of specific interventions, programs and choices, as well as project the effects of changes in program design and governance on the broader society. These estimates could be made available in a transparent and real-time fashion to all actors in the system, including users of programs. With access to this rich base of information, individual citizens will be empowered to choose for themselves from among various types of support when making big life decisions. This could lead to a major shift in service delivery: governments and service delivery agents currently have the dominant role in choosing the best programs or services for clients.
This kind of predictive analytics has already been widely adopted throughout the private sector – from reliance on “sabermetrics” in major league baseball (Lewis 2004) to the transformation of marketing and retailing by the behavioural science of microtargeting (Issenberg 2012; Mayer-Schonberger and Cukier 2013). Online shoppers see these methods in action when a website suggests goods and services based on their history of searches and purchases. Yet, at some level, the concept of predictive analytics is so far removed from the way government and social policy operate today that it sounds like science fiction.
Most current policy observers will likely find it odd, even perverse, to suggest that the national statistical system (which will collect the necessary data to generate the statistical estimates) would be called on to play such a central role in social policy. Today, we still mainly see the role of the statistical system as one of responding (often highly imperfectly) to the changing needs of policy analysts. That is, the statistical system continues to be seen as a largely passive player that speaks only when spoken to, rather than as the central resource for the data that drive program operations and governance. It may even replace programs by providing current information directly to citizens.
How Programs Are Changing
Through many subtle changes in policy and program design in recent years, Canada has been slowly moving, intentionally or not, toward this new paradigm of social policy: the enabling society. To place these changes in historical context, box 2 lists selected milestones in Canadian public policy, from the introduction and maturation of the welfare state to the beginnings of the enabling society. The shift is evident in four main areas, which are discussed in greater detail in this section:
- individualization of income security and the use of life-course income transfers
- evolution toward a guaranteed annual income
- individualization of employment support services
- tentative development of a “what works” evidence base
Individualization of income security and the use of life-course income transfers
As noted earlier, a key trend in social policy has been the gradual shift toward individualization in the financing of some social programs. Under an individualized model, savings are used to transfer a person’s income across that person’s own life course, between periods of high and low income. While this model does not remove the need for income security policies that redistribute income between different groups of people at a given point in time, the increasing preference for individual savings policies over direct income transfers does constitute a very different approach to the way resources are allocated and accessed within society. The shift toward individualized life-course income transfers began with reforms to the retirement income system and has continued with the adoption of asset-building approaches in other areas of social policy.
Individualization of pensions
This shift is already well under way in retirement income systems around the world. It has occurred through greater prefunding of benefits, greater emphasis on defined contribution programs, increased use of tax-advantaged individual retirement savings accounts and reforms to traditional defined benefit public pensions that more directly link individual -pension benefits and contributions. In some countries, as in Sweden, individual accounts have been built right into the main public pension system.
As a result, people are now more directly responsible for funding their own retirement, and they have more choice in big life-course decisions, such as the transition from work to retirement. Reforms to raise the eligibility age for access to old-age benefits and to allow people more flexibility as to when they start to receive those benefits have also worked in this direction.
Asset-based individual accounts
Over time Canada has developed many individualized, tax-deferred and tax-advantaged asset-building programs to address other social policy objectives. Examples include the Registered Education Savings Plan (RESP) to support individual savings for children’s education and the Registered Disability Savings Plan (RDSP) to help families support adults with disabilities. In both programs, governments supplement citizens’ own contributions, in recognition of the broader social policy objectives that are achieved by these targeted savings programs. Not all initiatives have specific policy targets, however: the Tax-Free Savings Account (TFSA) encourages people to save more and provides complete flexibility as to how the money can be spent. Although available TFSA contribution room is small today because the program is still fairly new, its long-term impact on aggregate savings is likely to be quite significant.12
Other extensions of this model have also been contemplated. A major Canadian experiment in the 2000s, known as Learn$ave, confirmed that, by matching people’s contributions, governments can encourage low-income adults to save for and enrol in education and training (Leckie et al. 2010). This project shed important light on how incentives can be used to increase savings among vulnerable populations, and also the level of incentives required to affect individual behaviour within these groups. Although individualized learning accounts have yet to be introduced in full, further experimentation has been done to examine their role in other settings (Ford et al. 2012).
The learning account model builds on a large body of evidence from the United States and other countries where individual accounts have been used to stimulate saving among low-income and vulnerable populations. Depending on the program, the savings can be used to support further education, start a business or purchase a first home. The programs are part of a movement often referred to as asset-based egalitarianism, which attempts to counteract the problems of income-based support, such as the disincentives to work and save. The theory behind the movement holds that control over one’s assets is an important element in control over the course of one’s life – a key factor in, for example, helping people move out of poverty or dependence.
Other countries have gone even further in exploring asset-based programs. The UK introduced an investment fund, the Child Trust Fund, which operated as a long-term tax-free savings account. Accounts were automatically established for each child at birth. The government deposited seed money, which could be supplemented family members and others. The children could use the savings in their adult years. The Netherlands started a tax-supported, voluntary individual life-course savings scheme in 2006. This program allows employees to take time off work for purposes of child care, elder care, educational leave, travelling, sabbaticals or part-time work when nearing retirement.
Initiatives such as these have generally produced mixed results. Some are large established programs, while others are little more than demonstrations and experiments. Some are successful, while others have not achieved their intended outcomes. We are still in the development phase. However, the general direction is clear: governments are moving away from their exclusive reliance on point-in-time income transfers and supplementing these programs with asset-based life-course-oriented approaches that are, by definition, individualized.
Evolution toward a guaranteed annual income
The first significant use of modern ICT to redesign the actual content of Canadian social programs was for the development of a series of income transfers within the tax system, through deductions and refundable credits, that amounted to a modified version of a GAI. A GAI has long been seen as an ideal public policy; some experts have even proposed replacing most income support programs with a simple, nonintrusive income transfer that would end, or virtually end, poverty (see Courchene 2001, 2012; Courchene and Allan 2009).
Since the welfare state was established in the 1960s and early 1970s, Canada has had a GAI for seniors in the form of the GIS and OAS. These two programs are the first pillar of the retirement income system. Major extensions of the GAI occurred with the introduction of the National Child Benefit (NCB) in 1998 and the Working Income Tax Benefit (WITB) in 2007. The former uses the tax system, along with other measures, to support low-income families with children. The latter provides earnings supplements for those in low-paying jobs. These innovations were made possible by using complex statistical models to analyze, test, design and integrate these measures within the personal income tax system.
Individualization of employment support services
While the transition to the enabling society is well advanced in the design of some income security programs, at the level of service delivery it has hardly begun. However, at least one pilot initiative in the delivery of active labour market programs (ALMPs, which include training programs, job creation, wage subsidies and job-search assistance) provides an example of how all social services may be transformed.
In the 1990s, Human Resources Development Canada (HRDC), the predecessor of Employment and Social Development Canada (ESDC), played a world-leading role in developing and piloting a highly individualized approach to the delivery of active labour market interventions. The initial pilot study created a longitudinal file drawn from some 19 provincial and federal administrative data sets, covering more than 10 million clients over nine years. The data were scrubbed in order to anonymously track clients against 250 variables and outcomes, using tax, EI and other administrative records. From this blended information, the department was able to estimate what would work best for various types of clients, in various regions.13 These estimates could then be used to make appropriate referrals to each of the 25 available ALMP interventions. The techniques that were used underwent a number of revisions based on lessons learned. Near the end of the decade, they were working sufficiently well that full-scale implementation would have been possible.
The system was designed so that employment counsellors could feed into the database information about the clients sitting in front of them, and a few seconds later a response about the type of ALMP intervention likely to work best for each client would pop out. The calculation of potential outcomes was based on a technique that matched each client with earlier participants who had similar characteristics and then examined those participants’ subsequent history of employment, earnings and receipt of social benefits across each of the interventions they had been referred to. The system could also produce for management purposes information about what kinds of interventions were working best overall and which were in greatest demand. Although crude by today’s standards, these outcomes-based calculations were still much more useful than any others being done at the time.
As it is now, privacy was a major concern in the early days of this pilot. Since that time, the development of newer techniques, such as synthetic cohort analysis, means that no information connected to actual, identifiable Canadians will ever be revealed or used in the process. All data are anonymized before calculations are performed.
The pilot was abandoned for a variety of reasons (Colpitts and Smith 2002), largely related to organizational restructuring and a change in policy direction within HRDC. However, this initiative could be readily reintroduced as part of Canada’s federal-provincial–territorial labour market agreements.
Tentative development of a “what works” evidence base
We do not ordinarily think of information and statistics as a separate policy instrument of social policy. Rather, they are seen mainly as tools of analysis in support of other instruments or programs. In the enabling society, information will be a major product of government, to be used both independently and in support of other policy instruments.
The social statistics systems of the future will be based on huge data warehouses drawn mainly from administrative files. These will allow microlevel assessments of what has worked in the past and what is likely to work best in the future for various clients. Today, however, the existing statistical system is still fundamentally structured in very traditional ways: survey by survey and census by census, using approaches that were developed long before today’s advanced technology. Most analysis is similarly based on separate, cross-sectional data sets. This approach made sense for the point-in-time designs of welfare state programs. However, our statistical base and our analytical tools are rapidly losing relevance as we move toward the enabling society.
Nevertheless, progress has been made in several areas, preparing the ground for the transition to big statistics. Notable developments include individual initiatives to link administrative data sets with census and labour force information, and efforts to develop new analytical models. Overall, however, progress has been slow, and in some areas, there has also been backsliding, particularly in labour market statistics (Drummond 2014).
Statistics Canada’s LifePaths microsimulation model is an interesting example of how a potentially important innovation has not been fully exploited. The model allows the life courses of millions of Canadians to be anonymously simulated using data drawn from many sources: administrative files, surveys and censuses (Statistics Canada 2013). For each of these synthetic individuals, it creates life histories from birth to death that are representative of the history of Canada’s population over the past half century but are not linked to or representative of any one person. LifePaths may not be the ideal tool of analysis; it is incomplete and, in some cases, too complex for many analytical tasks. However, LifePaths had many of the features that will eventually become part of the analytical tool kit in the enabling society.
For example, LifePaths shows how a big statistics approach can create new information that is vastly more powerful and detailed than any of its component data sources standing alone. The model centres on the overlapping life-course trajectories of individuals, including their employment, earnings and family trajectories. It situates those people in the family, social and economic contexts in which they actually live. It provides a highly detailed view of one individual’s life course but can also produce rich cross-sectional analysis when needed. It traces the interaction of individuals with government income programs and can examine the combined effect of changes in program design – including their interaction with other programs. And, importantly, LifePaths has predictive power that can produce the kind of “what works” data that will be commonly used to design and manage programs in the enabling society. It does all this within a framework that, being based on synthetic individuals, completely protects privacy.
However, the story of LifePaths suggests that we are only beginning the transition to the enabling society. It was developed some 20 years ago and, despite its enormous power, has been used surprisingly little since then. The underlying problem is that the model is designed to answer the kinds of analytical questions that are about the whole of society and the whole of government, questions that are hard to address in today’s welfare state. (I address the challenges governments face today in adopting a whole-of-society perspective later, in the section “Reshaping Governance.”)
LifePaths has also demonstrated the difficult challenges of securing financing to develop and sustain highly sophisticated models of its kind. Policy research and analysis are funded mainly in order to support current priorities; few resources are available for building the capacity to meet tomorrow’s needs. Statistical expenditures that lay the groundwork for future applications, such as longitudinal surveys and microsimulation modelling, often get chopped in times of budget cutbacks. Indeed, in December 2014, as this essay was nearing completion, Statistics Canada announced the termination of financing for LifePaths.14
What is needed is the leadership to empower Statistics Canada to develop a long-term road map for the statistical system that moves us beyond the old ways of doing business and allows us to produce the information and analytical tools required for the enabling society. Such a road map will need to be developed in close collaboration with data users, such as provincial governments and other federal departments. An action plan is discussed in the final section of this essay.
How Social Policy Will Change in the Next Decade
The individualization of income security and the increased use of life-course income transfers will have implications for other areas of social policy, most likely resulting in a system with three pillars:
- guaranteed annual income
- social insurance
- lifetime accounts
Each of these pillars would consist of many separate programs, originating in different departments and jurisdictions, that would have harmonized and consistent goals. The new system of big statistics would provide the complementary support necessary for integration and collaboration between actors.
Guaranteed annual income
The evolution toward a GAI is likely to continue in the enabling society. This form of income security would provide the assurance that a basic minimum income would be available at any time to people who have fallen through the cracks in other social programs and supports. For reasons of fiscal sustainability, as well as to avoid the potential for new disincentives to arise, the GAI would be set at a relatively low level. More importantly, a GAI should be a relatively small component of the whole system. Since it is ultimately based on an arbitrary poverty threshold derived from broad averages, a GAI reflects a point-in-time concept of exclusion that is at odds with the evidence referred to above about the duration of poverty spells. Nonetheless, a GAI will provide a critically important backstop, while the other pillars are more attuned to the unique needs, capabilities and circumstances of most individuals over the course of their lives.
The most straightforward way to fully establish this pillar would be simply to make additional funding available through the NCB Supplement/Canada Child Tax Benefit (CCTB), the WITB and other supports now provided by existing provincial social assistance systems. As well, a comprehensive refundable tax credit could take over the income support functions now provided to seniors through the OAS, the GIS and existing tax measures. Such a combined seniors’ benefit was proposed in the 1990s but not acted on at that time.
Social insurance would constitute the second pillar of income security in the enabling society. It includes programs to cushion citizens against unexpected disruptions in life such as illness, injury and job loss. Social insurance programs pool these risks and the government acts as the insurer. The main components of this pillar would be health insurance, workers’ compensation, the CPP/QPP’s disability insurance programs and the regular benefits component of employment insurance.
The transition to the enabling society is not likely, in and of itself, to bring about large changes to the design of social insurance programs, apart from separating out program functions that properly belong in the other pillars. Nevertheless, some indirect effects will be important, particularly for EI.
Many studies have documented serious problems in the way EI is designed and have proposed major changes to simplify the system and allow it to operate more like a traditional insurance program. Yet, despite significant consensus among think tanks, academic experts and various other commentators, changes to EI have been largely incremental, leaving intact the basic structure established in 1971.
The underlying problem lies in the gradual blurring of the distinction between insurable and noninsurable joblessness in the labour market. Social insurance makes most sense for experienced workers following traditional career paths who face a temporary period of unemployment. However, this segment of the labour force is shrinking when compared with others: Canadians in nonstandard or precarious work (such as temporary, part-time or self-employed workers); those whose jobs disappear when whole economic sectors undergo major restructuring; or those who need alternative work arrangements to deal with family responsibilities or important life events.
The obvious “clean slate” response would be to keep EI as an insurance program for the dwindling portion of the workforce for which it is appropriate and to develop separate interventions for those with more complex needs. Apart from the inevitable difficulty in drawing the line between these two groups, the biggest obstacle to such a solution would be the additional, noninsurance functions that EI has taken on in recent decades. These include parental and compassionate leave, funding for provincially delivered training and employment support programs and, indirectly, regional income redistribution. If the intent were to simplify EI, governments would have to find alternative ways of achieving these other goals.
The question is whether the shift to the enabling society will help solve this problem. Of course, no change in program design can make difficult political choices disappear. Any significant reform to EI is likely to create both winners and losers across regions, industries and occupations. That is unavoidable. Nevertheless, some key elements of the enabling society will make reform easier.
First, the creation of a separate lifetime accounts pillar (see below) would provide a new and more sensible home for some of the special benefits (such as compassionate care and parental leave) and employment training functions currently performed by EI. Second, and perhaps more important, big statistics will provide a much better understanding by all parties of the actual effects that changes to programs such as EI are likely to have, both in the labour market and on other related programs, including those at other levels of government. Evidence-based reform will not eliminate conflicts based on varying jurisdictional interests but it could help resolve them.
The biggest change to income security in an enabling society would come from the creation of the lifetime accounts pillar. Most existing pension and savings programs would fall under this pillar (including the CPP/QPP, RRSPs and RESPs), except for those primarily designed to provide basic incomes or disability insurance (such as GIS, OAS and the CPP/QPP disability insurance program). In addition, this pillar would take over elements from income support programs whose goal is to provide assistance with a major life event or to facilitate a long-term transition, such as the special benefit programs and training supports currently within EI. A number of tax credit and loan programs could also be integrated – the range of options is quite large.
The intent in establishing this pillar would be to provide individuals with more flexi-bility in accumulating the resources they will need to navigate the major stages of life and to do so in a way that is more coherent, transparent and straightforward than the mechanisms of many existing social programs. Although the goals are simple, the degree of change would be transformative.
While the lifetime accounts pillar would consist of different programs that work in a harmonized manner, it is easiest to understand if one thinks of it as a whole. Box 3 illustrates how it might operate.15 It is important to stress that this system would be introduced only gradually, following extensive research, development and consultations. In designing the lifetime accounts pillar, governments would retain maximum flexibility – the concept can be applied broadly or narrowly, and in accordance with whatever values are deemed appropriate.
The relative size of the pillars
The relative size of the three pillars would change over time depending on prevailing circumstances and values. In the initial stages, it is likely that only small changes would be made to existing programs in order to minimize the risk of disproportionately affecting certain groups, and to build confidence in the new system.16
It is likely that the lifetime accounts pillar would grow relative to the others, but this should happen gradually on the basis of evidence. In terms of those who stand to benefit most, that will depend on the policy choices of the government of the day. However, on balance across the three pillars, it seems likely that potential winners would include those with low-wage jobs (through an expanded WITB), families with children (through an expanded NCB Suplement/CCTB), those who choose to spread work and formal learning over a longer period of their lives, those who choose to retrain or make midcareer adjustments, and (for reasons discussed in the next section) those who face multiple obstacles to participating fully in society.
The enabling society will shape the design and delivery of services in important ways. These include individualization of services based on evidence of what works, greater integration of different types of service, a gradual change in the role of government, and a shift from a social expenditure to a social investment model.
Individualized and evidence-driven services
The tailoring of services to individuals, as piloted in the 1990s ALMP study discussed earlier, will eventually be available for all social services. Big statistics will process microlevel evidence about not only individual outcomes but also related costs, processes and outputs in order to provide balanced, consistent information supporting the decisions of all those who participate in the delivery system.
The speed of implementation is contingent on the availability of the microlevel evidence, however. Attempting to introduce this kind of model without that evidence would be harmful, raising expectations that could not be met. On the other hand, the evidence need not be complete or of uniformly high quality at the outset. It is quite possible to begin gradually in those areas of service delivery where partial evidence exists, provided this is done on an experimental basis and with realistic expectations.
A gradual approach would bring large payoffs. Feedback loops allow the quality of intervention design to improve over time, as we learn from the results of earlier experience. This learning can then be applied to other kinds of service delivery, which would in turn accelerate the transition to evidence-driven programming. The result would be vastly better than with the traditional “all at once” implementation strategies often preferred by government. Implementation could begin quickly in the case of ALMPs and other areas of employment services where intergovernmental labour market agreements give provinces and territories a lot of latitude to experiment and innovate.
Education is another domain that could be an early adopter. We are getting close to having the data needed to trace students’ educational and learning pathways and to gauge which pathways work best for whom. Standardized data on outcomes have been developed through the Council of Ministers of Education, Canada, leading to comparative studies of student performance at the elementary and secondary levels in many provinces and Canada’s active participation in new international assessments of literacy and numeracy. Data linkages are also being developed to integrate data from various PSE institutions with anonymized data derived from tax files, although this work is still at an early stage. Longitudinal surveys now allow us to measure the lifelong impact of such programs as early childhood education. While we are still far from the kind of consistent, detailed, longitudinal data on learning that would be needed to run a full-scale system of “what works” evidence, we are near the point where serious development work can begin.
Similarly, in the health care system, we are close to an evidence-driven system in some areas while far away in others. For example, compared with evidence about educational tools or ALMPs, there is much better information on which kinds of medical interventions work best, including information derived from clinical trials and related meta-analysis. The challenge is to get better information on life-course outcomes – in particular, by linking information about the medical interventions an individual receives at different points of life with socio-economic data and life-course characteristics.
In health policy, compared with other areas of social policy, there is also a much greater willingness to fund research and to tie policy to that evidence. For example, large sums have been spent on developing a rich digital information base on medical interventions. In Canada, an independent agency, the Canada Health Infoway, has made significant progress in working with many partners to digitize medical records. While that effort is primarily devoted to making the existing system better, the health databases being constructed should be very useful in developing evidence-based, individualized health services. We can only imagine similar possibilities in many other areas of social policy.
As services are increasingly individualized, a host of changes will follow in the organization and management of services. Not only will big statistics allow for the individualization of particular service interventions, it will also support a more integrated approach across different types of service interventions, whether they relate to employment, learning or health. Integration will take place particularly at the front end of service provision, where referrals are made.
Today, hundreds of thousands of separate interventions occur across the social support, education and health systems. While referrals are often made between these systems, they are not yet integrated enough to be able to share information effectively.
Backed by an integrated system of microstatistics that tracks all service interventions in a standardized way and provides “what works” data, service providers will be able to put together the most appropriate support package for each client, by better understanding how their own interventions interact with those delivered by other providers serving the same person. Services may still be delivered through separate entities, but the integration between service systems will be far more seamless. The resulting system will continue to learn and improve over time as it accumulates and analyzes ever more information.
Role of government
The lack of common, transparent information that can be shared among various systems and service providers has been a major constraint on individual choice and on the ability of government to leverage market forces as part of service delivery.
As the enabling society fills in these important gaps, the roles of governments and citizens and the interaction between them will change. Citizens will have access to much of the same information that service providers have, empowering them to become informed consumers. These changes could also enable governments to play a less active role in service delivery, allowing them to focus more on managing systems and service markets (that is, setting policies and regulations and providing the necessary information for market forces to operate). Governments are likely to remain the main funders of most programs and services, with the key difference being in how funding is delivered. Payments will increasingly be made through funding mechanisms that contract service providers (public, private or not-for-profit) to achieve particular outcomes, rather than paying providers on the basis of the number of clients served.
As the service provider community becomes more diverse, intermediary bodies at the community or regional level will develop further to help interface between governments and service providers. I will consider in the following section the issues of governance that this transformation raises.
From social expenditure to social investment
In most cases today, the funding of social programs is treated as a current expenditure, reflecting the point-in-time perspectives of the welfare state. Funding comes from current revenues and is accounted for as part of the annual budget cycle, without any specific relationship between the size and timing of the funding and the size and timing of future gains that might result from that expenditure.
With the adoption of life-course and asset-building approaches as part of the enabling society, a true model of social investment could be applied in budgetary planning. Spending decisions will be based on actual calculations of the expected future benefits from those investments, and accounted for as such. It is likely that, as part of this shift, governments will also allocate more spending into separate earmarked accounts so that investments can be managed more transparently.
Information as a policy instrument
At first glance, the kind of big statistics discussed in this essay seems like a completely different animal from the current system of national social statistics produced by Statistics Canada. The sources of the data are somewhat different: surveys and censuses will continue to be important, but their role will increasingly be seen as filling gaps in the new primary source of data which will come from administrative files in government departments. The users and uses of the data are also completely different. The audience for current national statistics is mainly researchers and analysts in government, think tanks and academia who use the data mainly for cross-sectional analysis. By contrast, big statistics will be focused on delivering information primarily to citizens, with essentially the same data also supporting a range of other actors who will use this information in the design, delivery and assessment of programs and services. For both sets of users, the goal is to inform decision-making in a timely manner.
The key implication of this change is that information will become a separate policy instrument of government. Box 4 illustrates how this dramatic shift is likely to play out, using the example of labour market statistics. A public website, perhaps building on the platform already in place through the federal Job Bank portal, could provide individually tailored information on what options are likely to work best given an individual’s specific circumstances. This information will help people make better employment and educational decisions; by extension, it will improve labour market efficiency by more precisely matching labour supply and demand.
The differences between the kind of information described in box 4 and the current weak system of labour market information appear so large that the development of big statistics may be on a separate track for some time from ongoing reforms by Statistics Canada. Later I will propose an action plan for the transition of the statistical system.
Placing information in the context of the user
As information evolves into a distinct instrument of social policy, it is important to recognize the context in which it can best be used. The credibility of the system depends on finding a simple way of informing users of the extent to which they should factor in the data when making decisions. This is not difficult conceptually, but it is seldom done well today. The evidence base will vary greatly in quality, at least initially. It will be important to communicate this initial variability to service providers and users. The quality of the data will improve greatly over time as more data and evidence are gathered and added to the system.
At the end of the day, personal judgment will remain the most important factor at all levels of decision-making – for governments, service providers and individuals. What will be different in the enabling society is that quantitative information to support those judgment calls will be vastly stronger, even in the early days and increasingly so as the system matures.
The availability of new types of data will have a significant and positive impact on how governments design and deliver services. However, better data will not, on their own, bring about the policy transformation we need if we do not also address the severe governance challenges facing governments today. Chief among these are the gaping holes in how accountability is understood and supported in the context of both vertical (within a program or department) and horizontal (across departments and jurisdictions and in partnership with the groups and individuals being served) management. If the shift to the enabling society is to lead to greater collaboration, integrated service delivery and life-course planning, then accountability structures (at all levels) will need to be transformed as well.
The key challenge for government accountability is to measure the performance of programs and services in a way that reflects the complex and interconnected environment in which many of them operate. Evidence-based accountability is lacking for several reasons.
- There is no single logic model that meets all needs. Governments measure performance and accountability using what are known as logic models. These models attempt to map how inputs and processes lead to outputs and, eventually, to the outcomes targeted by a particular program or service. The logic model used by one program may or may not mesh with the models applied to others that serve the same type of clients. What’s worse, the typical logic model used in government stops at the level of a specific program or department. This siloed approach is inadequate for measuring outcomes, since outcomes frequently involve a number of vertical and horizontal relationships as clients interact with various programs, processes and entities across time. Someone who is unemployed and seeking a new job will likely use a variety of services in that journey, including direct supports such as EI or welfare, training and job-search assistance; indirect factors such as child care, transportation and literacy will also make a difference. Failing to examine all these relationships from a whole-of-government or whole-of-society perspective makes it difficult to attribute high-level outcomes to the particular activities of a specific program or service provider.
- Outcomes are ambitiously set but poorly measured. Governments seek to manage on the basis of outcomes, often expressed in lofty terms: for example, “a skilled, adaptable and inclusive labour force and an efficient labour market” (ESDC 2014). Typically, good data are available on inputs (such as costs) and outputs (such as number of clients served), but there are big gaps in the evidence documenting outcomes over the medium and long terms. As a result, outcomes are frequently measured using indirect, point-in-time indicators (such as the unemployment rate) that give only a partial view of how programs contribute to changes in society.
- Stakeholders have diverse preferences. Particularly in the area of social policy, service delivery must accommodate the needs of a multitude of stakeholders, service pro-viders and individual citizens who use programs. Each of these actors has different preferences, values and expectations with regard to policy, a diversity that is often too complex to incorporate into accountability structures. Governments are then forced to simplify outcomes using abstract or generic benchmarks that do not reflect actual complexity or evidence.
Any effort to put in place a system and a culture of evidence-based policy-making must first get past these obstacles. Success will come more easily, however, once the major transitions described in this essay are under way. The most important of these is the overhaul of our statistical system to help us better understand the complex pathways that people take over the course of their lives. Big statistics will radically improve the measurement of outcomes and our ability to use that information to measure not only the effects of programs but also the way in which they are delivered and managed.
Moving to better evidence-based decision-making
The system of big statistics envisioned by the enabling society will allow us to rebuild our logic models from the ground up. Rather than picking broad indicators (such as the unemployment rate to measure success at improving the efficiency of the labour market), the key is to start by using microlevel data to tell us how outcomes actually occur and then design accountabilities around those results. To achieve this result will require major developments in data collection and analysis at several levels.
- How citizens use programs: Better data would show how the interventions people use interact with and produce outcomes related to the events and activities that occur in all domains of life – work, family, the community – and to participation in other government programs.
- How interventions are provided: Going beyond the traditional measurement of inputs and outputs, governments will also need to gather information on the specific techniques used in different programs and services, as well as about other elements that help to define how each provider offers its services.