Ottawa should invest in an arm’s-length agency to collect data on the well-being of Canadians and measure the outcomes of government policies.

How many Canadians relied on social assistance last year? How many experienced food insecurity? Was there an increase or a reduction in long-term poverty? The truth is we don’t have the answers to any of these questions because quality data simply do not exist. And yet, without them, making policy to reduce poverty begins to resemble throwing darts in the dark.

In his budget speech on March 22, Finance Minister Bill Morneau presented the Canada Child Benefit as “the most significant social policy innovation in a generation,” with an estimate that it will lift 300,000 children out of poverty. The budget also trumpeted a commitment to evidence-based decision-making. As Morneau said, “If we are to lift children out of poverty, we must first understand the cause.”

But, in fact, we lack the data to do that. The federal government still has not connected the dots between major social expenditure and the information required to guide and evaluate social policy.

To bridge this divide, the federal government ought to direct some of the $3.4 billion committed to social infrastructure to fixing the data infrastructure that is essential to identifying need, measuring progress and developing evidence-based social policy.

One way to direct such an investment would be the creation of an arm’s-length agency to report on Canadians’ well-being and social policy outcomes — a Social Indicators Observatory (SIO).

Over the past decade Canada’s social policy information infrastructure has crumbled. While the cancellation of the long-form census was the highest-profile example of the head-in-sand approach, dozens of ongoing surveys were also cancelled, such as the Survey of Labour and Income Dynamics (SLID).

Currently we lack timely, reliable indicators of how Canadians are faring: Statistics Canada income data are released with a two-year delay, and we have no regular and longitudinal national measures of health, wealth and material deprivation. The disbanding of the arm’s-length National Welfare Council in 2012 ended the reporting of provincially comparable information about social assistance rates and caseloads. Nonprofits such as the Caledon Institute and the Child Care Resource Centre have been driven to crowdfunding for the collection and distribution of key information on Canadian social policy provision — data that were previously federally funded.

The establishment of the SIO would also lay a solid foundation for delivery on another federal government promise: a national poverty reduction strategy. Key components of poverty reduction strategies are the setting of targets and tracking of policy performance. However, with very limited data on how Canadians are doing, the potential of these strategies is also limited. For instance, Saskatchewan’s Poverty Reduction Strategy, released last month, is based on poverty statistics from 2013. It has only one target (rate of persistent poverty), for which there are no publicly available data.

On their own, few provinces or territories have the capacity to collect and analyze socio-economic indicators. Aside from the inefficiencies of separate collection and gaps in coverage, provincial data are often not available publicly or in a format that allows interprovincial comparison. As a result, provinces cannot easily learn what works from one another. Nationally, we don’t have a clear picture of our social architecture or its outcomes.

The SIO’s mandate would be to foster policy learning through the provision of data to analyze progress on social goals and policy outcomes. It seeks to fill what is an inherently intergovernmental function, and, as such, it requires provincial buy-in. To get the provinces to play ball will require the right mix of federal cash, an acceptable mandate and political leadership.

Making provincial data comparable means facing some politically unpopular facts: for instance, finding out which provinces have the highest social assistance rates or the largest unemployment traps. Provincial governments must be enticed to look beyond these inevitable embarrassments, to the power of leveraging federal dollars and statistical expertise to make more informed social policy choices. The era of ostrich-style policy-making must be over.

We envisage the SIO engaging in three types of activities: producing indicators, reporting on social policy and supporting research and collaboration.

The core business function of the SIO would be regularly collecting or collating data to provide indicators of Canadian well-being and social policy outputs. Here we imagine bringing together data from Statistics Canada, such as low-income rates, alongside surveys of material deprivation and subjective well-being. An important contribution would be harmonizing provincial data (including administrative data) to give comparative and national metrics on policies under provincial jurisdiction, such as social assistance, child care and minimum wage.

Alongside national indicators, the SIO could be tasked with reporting indicators requested by individual provinces. Currently, the provinces have to pay Statistics Canada to get customized data to monitor their poverty reduction strategies. Reducing this expense would make the SIO an attractive endeavour to take part in. In this core statistical function, the SIO could be compared to the European Union’s statistical agency, Eurostat. It would offer visible and independent accountability for government policy-making at all jurisdictional levels.

A second role of the SIO would be to regularly report publicly on these indicators and provide basic analysis of trends in social policy and outcomes. In this function the SIO could provide the value-added functions of the OECD, which not only delivers comparable data but provides case studies and best-practices research.

The third function of the SIO would be to act as a hub for interprovincial learning and social policy research, as well as supporting province-specific information needs. Sharing the SIO data would allow governments, academics and research institutes to advance research on well-being and public policy in Canada. The SIO could also provide support to individual provinces on their specific data and policy evaluation needs — allowing the federal government to support provincial autonomy while realizing economies of scale.

The SIO’s biggest potential benefit would be what it could do for Canadians by guiding sound social policy-making. A strong social data infrastructure will contribute to real progress on poverty reduction by offering consistent and comparable performance data to inform governments and establish best practices. While this may be most critical to those living in poverty, it will also serve the millions of middle-class Canadians who are one divorce, birth, illness or job loss away from significant hardship.

For the Trudeau government, the Social Indicators Observatory would put principled words into action. Moving forward from the reinstatement of the long-form census, it would show commitment to this statement in the budget: “By focusing on outcomes for Canadians and making evidence-based decisions that are anchored in meaningful data and indicators, the Government is moving to a culture of measurement and impact, and is putting in place the tools to deliver on priorities, align resources to programs and activities that deliver real value for Canadians, and provide meaningful information to Canadians and Parliament.”

Data gathering and dissemination may seem mundane tasks. However, much like the upkeep of bridges and highways, they provide a critical infrastructure for successful, functioning governance. Alongside investments in child benefits and affordable housing, it’s time the federal government invested in information that provides the foundation for a strong social infrastructure.