10 numbers we need in Ontario
High quality, timely data are essential for policymaking. The Institute works to enhance prosperity in Ontario and Canada, but needs access to data to put forward evidence-based analysis and recommendations to help with decision-making. On these 10 numbers, Ontario is at a decision-making disadvantage compared to other jurisdictions that do have access to them. Without adequate data, policymakers cannot fully understand a problem or opportunity, and cannot make the highest quality recommendations.
Here are 10 numbers we need in Ontario to improve policymaking:
Employment by cluster
To better understand clusters, data on employment and wages are desperately needed. This is especially at the local geography level (CMAs and CAs) and the 6-digit industry level. For instance, we do not know exactly how many employees work at a particular firm. As a result, estimates are used to determine employment and limit the effectiveness of cluster policy. In addition to firm size, there are limited wage data which means we cannot tell which clusters are more productive than others. Gaining access to this data will also allow us to make direct comparisons between Canada and the United States and is needed for effective cluster policy.
Policy area to be impacted: Cluster policy
Ontario lacks timely data on house prices at the neighbourhood level. Although widely known that there is a housing price bubble in Toronto, data are lacking on whether the overheating is due to market conditions such as population growth, immigration, lack of supply, or other factors instead of the presence of foreign buyers. Without this kind of information, policies designed to cool the housing market will be guesses at best. The province has started collecting information about buyers and it must continue to do so.
Policy area to be impacted: Housing regulations
Details on how firms in Ontario conduct training – and what kinds - are largely unavailable. Existing studies are survey-based with a high degree of selection bias. Without accurate information on how firms currently train and what their needs are, policymakers are left unsure how to design appropriate incentives or programs to meet the changing needs of the workforce. The Workplace and Employee Survey administered by Statistics Canada was an excellent resource for policymakers, but was last released in 2009. Surveys like this one should be reinstated.
Policy area to be impacted: Canada-Ontario Job Grant
- Adult skills
What skills do adults have? This question is difficult to answer. Education is typically used as a proxy for skills, as is occupational tasks, but this is not detailed enough to understand the direction of jobs or the labour market. While the OECD conducts some surveys on skills, Ontario should take a more active approach to finding out the skill levels of its individual. Canada needs an equivalent of O*NET OnLine – a tool in the United States for detailed data on occupations, skills, tasks, and education.
Policy area to be impacted: Employment Ontario retraining programs
- Number of patents
Patent data are one of the most widely used measures of innovation for an area. Provincial Canadian patent counts metadata are only available online going back two to three years. Improving data on patent counts in the province would give policymakers a better picture of innovation across Ontario’s regions.
Policy area to be impacted: Research and development tax credits
- Household assets and liabilities
Ontario needs a better picture of wealth inequality including debt loads and sources of debt. For example, how much impact does student debt have on the net worth of different age groups? This information is crucial for determining whether we wish to tax wealth, by providing estimates. Surveys by Statistics Canada fail to sufficiently capture detailed household assets and liabilities data at the provincial and local levels and are too infrequent.
Policy area to be impacted: Tax policy
- Firm records
Access to firm-level data, including corporate tax records, would provide opportunities to understand Ontario’s business climate in a new way. Researchers could better understand employees and wages tied to each corporation. This data would be useful to answering many questions related to the effectiveness of scientific research and experimental development credits (SR&ED), research and development credits, and the small business tax deduction. This could also feed into clusters research by providing specific detail about employment, geography, and profitability data.
Policy area to be impacted: Regional development policy
- Exports at a local level
Exports are a crucial indicator of the health and competitiveness of a region. Currently, export data is not available at the census metropolitan area level for Ontario. Provincial trade performance masks the differences in the industry mix and overall export performance across different metropolitan areas. In the United States, where metropolitan export data is available, the overall decline in exports in 2015 masked the strong performance in the largest metropolitan areas. Detailed export data is also only available for goods and not services.
Policy area to be impacted: Local economic development support
- Outcomes of government programs
To ensure policymaking is improving over time, evaluating current and past government programs is key. How can we learn from our mistakes if we do not assess the results? Much of this administrative data is kept within the government. The Open Data initiative is a step in the right direction, but much of the data is still ‘under review’ or ‘to be released’. Our hope is that internal government reviews of programs will be made open source as well.
Policy area to be impacted: All government programs
- A shop to make more numbers!
Ontario should unleash data science by establishing a Centre for Data Science for Social Good and Public Policy, similar to that of the University of Chicago. This body would work with non-profit and government by using big data and analytics to solve public policy problems. This will allow us to use the tools of modern data science to uncover the numbers, outcomes, and predictions that are currently just beyond our grasp.
Written by Julia Hawthornthwaite
Photo credit Ukususha