## cr17211

## Source details

**Canonical URL:** [cr17211](https://www.imf.org/-/media/files/publications/cr/2017/cr17211.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2017/cr17211.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2017/cr17211.pdf.json)

---

### C. Canada Infrastructure Bank (CIB): proposed design and fiscal features
- Proposed allocation: Can$35 billion (approximately 1¾ percent of 2017 GDP) over an 11-year period.
- The CIB will add to, and not replace, existing financing methods, including the federal government’s Can$187 billion Investing in Canada plan covering 12 years.
- Main objective (draft legislation): leverage private sector and institutional investor capital in infrastructure projects in Canada or partly in Canada that will generate revenue and that will be in the public interest.
- Financial and operational features:
  - Can$15 billion of the Can$35 billion authorization is expected to be expensed.
  - The CIB will be a wholly government-owned Crown corporation, subject to the Financial Administration Act (FAA) requirements (corporate plan, operating budget, capital budget for Government approval).
  - The CIB will be on-budget. Funding pursuant to the Minister of Finance’s statutory authority under the proposed CIB Act.
  - CIB borrowing is not currently envisaged; borrowing would require a borrowing plan in addition to the corporate plan.
  - The CIB and its investments will be on the federal government’s balance sheet; infrastructure-related special purpose vehicles (SPVs) in which the CIB invests will not be on the government’s balance sheet.
  - Explicit fiscal risks are limited to expenses reflecting the concessional nature of CIB investments or realization of associated risks, and to guarantees, which require approval of the Minister of Finance.

### C. Attracting private capital — mechanics and instruments
- Private investors seek financial returns, not economic or social returns.
- Key determinants of private investment (elementary finance equation):
  - Future value, or free cash flow (FCF): maximum financial returns available to private investors over the life of the investment.
  - Risk-adjusted rate of return = risk-free rate + risk premium.
  - Private investment = present value of future FCFs discounted by (risk-free rate + risk premium).
- Ways to increase private capital attracted:
  - Increase FCF:
    - Charge user fees to increase cash inflows.
    - Decrease cash outflows via private investor expertise across project lifecycle (design, construction, operations, maintenance).
    - Example: choosing a slightly more expensive design that significantly reduces operation and maintenance costs.
  - Reduce risk premium:
    - Diversify across projects or pool projects.
    - Shift early-stage project development responsibilities from SPV to CIB (permitting, zoning, public outreach).
    - Ensure predictable FCF via stable tax/regulatory environment; CIB guarantees of minimum demand, fee rates or total revenue; fee-rate adjustments by a trusted third party; equity investment by CIB viewed as partner.
    - Match project profile with investor knowledge through CIB staff market expertise and investor selection procedures.
- Brownfield investments are lower risk (development completed, known maintenance and demand). Recycling assets can free capital for new investment but may require users to pay for longer periods or higher rates.
- CIB as filler of funding gap:
  - CIB investment = total project funding required − private investment determined by discounted FCF.
  - As discounted cash flows grow, funding gap and CIB investment shrink.
  - Actual FCF exceeding amounts needed for private returns should be distributed to the CIB.
- Investment structure and allocation rules:
  - Minimum private investment required for CIB participation not officially announced; lower private participation approximates PPPs.
  - To keep SPV majority-owned by private investors, any CIB investment > 49 percent equity would typically be loans; permissible forms beyond 49 percent of total project cost should be determined.
  - Financial return allocation: FCF allocated first to private investors, then to CIB for return of capital and financial returns; CIB may accept lower returns to subsidize or be subordinate to private investors.
  - Estimated value of CIB risk mitigation efforts (guarantees, non-financial staff contributions) should be reflected in total project cost regardless of recovery provisions — raising valuation challenges.
- Investor rate-of-return discovery:
  - Ex-ante risk-adjusted rates are not known precisely and vary with investor familiarity.
  - Competitive bidding procedures should reveal market risk-adjusted rates.
  - Actual private returns during project life can be controlled via FCF allocation between private investors and the CIB.

### D. Effects on public finance, objectives, and measurement challenges
- Draft legislation’s objective of attracting private capital is general and needs refinement to inform measurement and trade-offs.
- Issues to address for public finance implications:
  - How to measure maximization of infrastructure spending?
  - How the CIB spends allocated money?
  - How budget funds are allocated to the CIB?
  - Fiscal risks arising from the CIB and its investments.
  - Effect of the CIB on the efficiency of public investment.
- Possible objective refinements:
  - Maximize total infrastructure spending by maximizing public and private financial resources.
  - Maximize total infrastructure spending under constrained public spending or borrowing.
  - Maximize infrastructure built by lowering project costs so the same spending purchases more infrastructure.
- Allocation and fiscal treatment specifics:
  - The CIB would be on-budget.
  - Within the proposed Can$35 billion, the Minister of Finance will allocate funds as needed under the proposed CIB Act.
  - Draft legislation is silent on treatment of return of CIB equity investments or retention of earnings.
  - Draft legislation is silent on whether approval in the budget process of CIB investments is in the aggregate or on a project basis.
  - SPVs in which the CIB invests would likely not be on-budget.
- Fiscal risks and transparency:
  - Fiscal risks are mainly implicit. CIB expenses include investment losses that lower government net worth without an unexpected demand on spending.
  - Guarantees require Minister of Finance approval under the FAA.
  - Corporate veil protects the CIB from SPV liabilities, but lack of SPV account publication could leave the federal government implicitly liable to finish/operate projects if SPV fails.
  - Confidentiality may prevent publication of financing terms for SPVs.

### E. Operational and prudential cautions; governance and public concerns
- Operational cautions:
  - Limit number and potential payout of guarantees to contain contingent liabilities.
  - Valuation challenges in disclosing and reflecting the value of CIB risk mitigation efforts in total project costs.
  - Explicit fiscal risks tied to concessional investments and guarantees requiring Minister approval.
- Organizational implications:
  - The CIB is better characterized as a fund than a bank and cannot be financially self-sustaining.
  - Clear division: planning process should evaluate social rates of return and identify project pool; CIB should focus on financial issues integral to its strategy and advance projects screened for public interest.
  - CIB staffing should reflect strategy:
    - Emphasize government/public service skills if focus is on non-financial risk reduction and interaction with governments.
    - Emphasize capital markets skills if focus is on creative financing and matching investor needs.
- Possible public concerns to address proactively:
  - User fees: public resistance likely; revenue from infrastructure is a key distinguishing characteristic of the CIB.
  - Perceptions of excess returns to investors: investor selection must be open and competitive to demonstrate low possible rates given project risk.
  - Subsidizing private investors: lower-than-market returns, write-offs, or grants may be characterized as subsidies though they may serve public interest if alternative is full government funding.
  - Input on project selection and implementation: concerns over geographic/jurisdictional distribution and local government voice in design/implementation.
- Communication imperative:
  - Communicating CIB objectives, policies and practices (including need for user fees) is critical to gain and maintain public support.

### Cost of funds perspective and strategic deployment of private versus public financing
- Federal government bond yield: about 1.9 percent annually.
- Risk-adjusted rates likely required by private investors: between 5 and 15 percent (assuming private investors bear at least some project risk).
- Implication: the same project FCF could support more government borrowing than investor capital.
- Recommendation: to minimize financing costs, government should borrow to the maximum extent possible within a prudent fiscal framework before using private capital—unless public borrowing is constrained.
- Contextual constraint: federal government wishes to limit borrowing and embark on a declining path of debt-GDP.
- Infrastructure demand estimates range widely from Can$150 billion to Can$ 1 trillion, implying government alone cannot meet demand.
- Tension between financial and social returns; role of user fees:
  - Charging user fees can reduce social returns if infrastructure is used at less than capacity.
  - If social benefits exceed financial returns and private returns insufficient, public subsidies are preferable until capacity is reached; prices/user fees should then be used largely for demand management.
  - Charging user fees could lead to unduly low demand and make society worse off — rationale supports CIB funding at less than market rates.
- Potential offset from private expertise:
  - Private expertise could lower project costs or extend asset life, offsetting higher financing costs.
  - Joint equity investment and life-cycle contracts with SPV can align incentives to lower total costs.
  - CIB should document extent to which private implementation reduces total costs relative to government implementation.

### Objectives for CIB deployment (implications for FCF and risk)
- Objective: Maximize the total amount of private capital leveraged
  - Advance by maximizing number of CIB projects and selecting projects with ease of implementation.
  - Encourage private investors latitude to select attractive (often lower-risk) projects.
  - Effect: maximizes FCF, minimizes aggregate CIB investment, faster implementation.
- Objective: Minimize the risk-adjusted rate of return required by private investor
  - CIB investment in a project is minimized.
  - Incentive for CIB to maximize non-financial methods to reduce investor risk (e.g., CIB staff handling early development).
  - CIB would want wide latitude to choose from a large project pool.
- Objective: Advance projects where private management expertise achieves significant economies
  - Favors specialized firms, large/complex projects, and projects benefiting from arm’s-length relationship from public pressures.

### Project selection strategies (three stylized strategies)
- Strategy 1 — maximize infrastructure built:
  - i) maximize annual mobilization of private capital (quantity and speed);
  - ii) FCF is maximized (user fees high, costs reduced);
  - iii) CIB takes relatively large amount of risk to lower investor risk-adjusted return and maximize leverage;
  - iv) large pool of projects available;
  - v) investors likely choose relatively simple, low-risk projects.
- Strategy 2 — maximize social rate of return:
  - i) large CIB equity investment at less than market rates;
  - ii) FCF lower than potential (user fees not maximized early), CIB covers higher share of funding;
  - iii) CIB takes relatively large amounts of risk;
  - iv) planning process important to identify highest priority projects;
  - v) investor selection adheres closely to planning priorities and requires government review of innovations affecting social return.
- Strategy 3 — maximize marginal benefit of CIB funds:
  - i) leverage most private capital per CIB dollar;
  - ii) emphasize non-financial means to lower investor risk (CIB staff assisting early development);
  - iii) FCF is maximized (user fees relatively high);
  - iv) large project pool to find highest return projects;
  - v) encourage private investor innovation to lower upfront cost rather than long-term cost or effectiveness.

### Box 1 — Social versus financial returns (analytical summary)
- Framework:
  - Vertical axis: private rate of return.
  - Horizontal axis: social rate of return (includes externalities).
  - Projects with positive externality lie to the right of the 45-degree line.
  - Private sector undertakes projects if private rate of return exceeds cost of funding (Regions A, B, and F).
  - Government prefers projects in Regions A, B, and C; Region C: high social return but private return insufficient.
- CIB role:
  - Government can provide financial incentives to raise investor risk-adjusted return (upward shift from X to X’).
  - Operational challenge: incentives should not be too generous; otherwise private return after incentives could exceed social return (point X”).
  - Requires fair and transparent competition and professional expertise in selection and negotiation.

### Comparative advantages of the CIB
- Two unique features among Canadian public infrastructure institutions:
  - Only the CIB is directed to invest in revenue-generating infrastructure.
  - Only the CIB is directed to invest equity alongside private investors.
- Equity partnership benefits:
  - Aligns interests of equity owners.
  - Enhances monitoring by providing access to SPV operational and financial information.

*Prepared by David Gentry (FAD) and Kotaro Ishi (WHD), excerpt from the IMF selected issues paper on Canada (June 21, 2017).*

---

### ESTIMATING THE GROWTH EFFECT OF PUBLIC INFRASTRUCTURE: evidence and empirical results

#### A. Context and policy background
- Federal commitment: Can$ 187 billion (equivalent to 7½ percent of GDP) in public capital; Can$ 96 billion is new additional infrastructure spending.
- This will triple investment in public capital over the next decade.
- Priority projects: public transit infrastructure; green infrastructure; social infrastructure (affordable housing, child care spaces, community centers).
- Objective: estimate macroeconomic effects of infrastructure investment using panel data for 10 Canadian provinces over 1961–2015.
- Key preview: public infrastructure investment could boost economic growth, complement private investment, and generate positive spillover effects across regions.

#### B. Size of the problem and trends
- Infrastructure gap estimates range from Can$150 billion to Can$ 1 trillion.
- World Economic Forum: Canada’s overall infrastructure ranking dropped to 21st from 17th a decade ago.
- Stock of economic infrastructure declined from about 20 percent of GDP in 1961 to as low as 12 percent in 2007; current stock is 5¾ percent of GDP below the 1961 level.
- Economic infrastructure defined: railways, ports, airport, pipelines, roads, and energy networks.

#### C. Empirics — data, model, and estimation
- Data: balanced panel of 10 Canadian provinces, annual data 1961–2015.
- Production function: Cobb-Douglas with infrastructure capital included:
  - Y_it = A L_it^{β_L} K_it^{β_K} G_it^{β_G} ḠO_it^{β_GO}
  - Baseline log-linear model: y_it = β_i^L l_it + β_i^K k_it + β_i^G g_it + β_i^{GO} ḡo_it + u_it; u_it = α_i + γ_i ω_t + ε_it.
- Two weighting methods for ḠO_it: geographical distance weights and simple average across provinces.
- Estimators used: Fixed effects; Mean Group (MG); Mean Group with a common dynamic process (MG-CDP).

#### D. Preliminary observations
- Full-sample scatter: fitted line y = 0.86x + 1.2, R² = 0.5 for average infrastructure capital growth rates vs average GDP growth rates across provinces (1961–2015).
- Literature context: an unanticipated 1 percentage point of GDP increase in investment spending increased level of output by about 0.4 percent in the same year and by 1.5 percent four years after the shock.

#### Regression results — key numeric findings (preserve reported values)
- Model summaries (selected coefficients preserved exactly):
  - Fixed-effects Models (1–3):
    - Model 1: G = 0.0448 (positive but not significant).
    - Model 2: G = 0.0463; GO (weighted average) = 0.0470 (both positive but not significant).
    - Model 3: G = -0.0296; GO (simple average) = -0.937 (negative and insignificant).
  - Mean Group (MG) Models (4–6):
    - Model 4: G = 0.0444 (positive but not significant).
    - Model 5: GO (simple average) = 0.0974 (positive but not significant).
    - Model 6: GO (weighted average) positive but not significant.
  - MG-CDP Models (7–9):
    - Model 7: G = 0.0768* (positive and significant).
    - Model 8: G = 0.0626 (positive but not significant); GO (weighted) negative and insignificant.
    - Model 9: G = 0.105+ (positive, significant at 15 percent); GO (simple average) = 0.274*** (positive and highly significant).
- Selected model-specific coefficients:
  - Real private capital stock (K):
    - Model 1 = 0.252***; Model 2 = 0.255***; Model 3 = 0.249***; Model 4 = 0.348***; Model 5 = 0.341***; Model 6 = 0.321***; Model 7 = 0.101***; Model 8 = 0.0836**; Model 9 = 0.104***.
  - Employment (L):
    - Model 1 = 0.750***; Model 2 = 0.748***; Model 3 = 0.734***; Model 4 = 1.355***; Model 5 = 1.173***; Model 6 = 1.155***; Model 7 = 0.560***; Model 8 = 0.507***; Model 9 = 0.526***.
- Memorandum items:
  - Observations = 550 for all models.
  - R-squared (Fixed effects models 1–3) = 0.988.
- Preferred specification and elasticities:
  - MG-CDP considered most plausible.
  - Models 7 and 9 suggest elasticity of infrastructure assets ≈ 0.08 (Model 7) and ≈ 0.10 (Model 9).
- Marginal product of infrastructure (MPK):
  - MPK computed as estimated elasticity × output-to-public-infrastructure ratio, based on elasticity 0.08 (Model 7).
  - MPK rose between early 1960s and mid-2000s as stock declined; MPK began to fall after the 2008–09 crisis as government investment rose, but current MPK remains above historical levels.
  - Policy implication: historically low current interest rates and elevated MPK suggest room to increase public infrastructure while keeping returns above cost of funding.
- Crowding-in (interaction terms):
  - g*k (own infrastructure × private capital): 0.00839*** (Model 1).
  - go*k (other provinces’ infrastructure × private capital): 0.0134*** (Model 2).
  - Interpretation: public infrastructure complements private capital; evidence of crowding-in.

#### Conclusions and policy implications from the empirical analysis
- Empirical findings:
  - Higher public infrastructure is correlated with higher economic growth.
  - Infrastructure investment can crowd in private investment (g*k = 0.00839***; go*k = 0.0134***).
  - Infrastructure investment has positive spillover effects across provinces (GO coefficients positive and in some specifications highly significant).
- Policy recommendations implied:
  - Improve coordination across jurisdictions to maximize economic benefits of infrastructure investment.
  - Given past underinvestment and historically low cost of funding, investing in infrastructure would generate a net positive return.
- Elasticities and key quantitative pointers:
  - Elasticity of infrastructure assets ≈ 0.08 (Model 7) and ≈ 0.10 (Model 9).

*Source: cr17211 (PDF chapter/section).*

### 1. Social Versus Financial Returns to Investment ________________________________________ 15

### 1. Social Versus Financial Returns to Investment

### A. Introduction and Context
- Infrastructure raises labor productivity growth and supports export performance; IMF research finds robust evidence that increasing public investment and the stock of public capital is positively correlated with economic growth.
- The World Bank and IMF jointly developed the Public Investment Management Assessment (PIMA) to promote public financial management (PFM) institutions that, taken together, result in more productive public investment.
- Countries are exploring institutions to mobilize private capital for infrastructure, including infrastructure banks (examples cited: Australia, the UK, state-level US, European Investment Bank).

### B. Canada Infrastructure Bank (CIB): Proposed Design Highlights
- Proposed allocation: Can$35 billion (approximately 1¾ percent of 2017 GDP) over an 11-year period.
- The CIB will add to, and not replace, existing financing methods, including the federal government’s Can$187 billion Investing in Canada plan covering 12 years.
- Main objective (as stated in draft legislation): to leverage private sector and institutional investor capital in infrastructure projects in Canada or partly in Canada that will generate revenue and that will be in the public interest (examples: supporting conditions that foster economic growth or contributing to the sustainability of infrastructure in Canada).
- Financial and operational features:
  - Can$15 billion of the Can$35 billion authorization is expected to be expensed.
  - The CIB will be a wholly government-owned Crown corporation, subject to provisions of the Financial Administration Act (FAA), including the requirement to prepare a corporate plan, operating budget, and capital budget for Government approval.
  - The CIB will be on-budget. Funding will be provided pursuant to the Minister of Finance’s statutory authority under the proposed CIB Act.
  - CIB borrowing is not currently envisaged; if it borrows, it would need to submit a borrowing plan in addition to its corporate plan.
  - The CIB and its investments will be on the federal government’s balance sheet; however, infrastructure-related special purpose vehicles (SPVs) in which the CIB invests will not be on the government’s balance sheet.
  - Explicit fiscal risks are limited to expenses reflecting the concessional nature of CIB investments or realization of associated risks, and to guarantees, which require approval of the Minister of Finance.

### C. Attracting Private Capital: Mechanics and Considerations
- Private investors seek financial returns, not economic or social returns. Social returns are effects on society not reflected in user prices; positive social returns are assumed when referenced.
- Key determinants of private investment in a project (per the elementary finance equation):
  - Future value, or free cash flow (FCF): defines maximum financial returns available to private investors over the life of the investment.
  - Risk-adjusted rate of return for an investor:
    - Risk-free rate (proxy: highly rated government treasury bills).
    - Risk premium: additional return demanded to compensate for project-specific risks.
- Private investment equals the present value of future FCFs discounted by (risk-free rate + risk premium). The present value rises with higher FCF or a smaller risk premium.

- Ways to increase private capital attracted:
  - Increase FCF:
    - Directly increase cash inflows (e.g., charging user fees).
    - Decrease cash outflows via private investor expertise across the project lifecycle (design, construction, operations, maintenance).
    - Example: choosing a slightly more expensive design that significantly reduces operation and maintenance costs.
  - Reduce risk premium:
    - Diversification across projects or pooling projects.
    - Shift early-stage project development responsibilities from SPV to CIB (permitting, zoning, public outreach), leveraging CIB neutrality.
    - Ensure predictable FCF through:
      - a stable tax and regulatory environment;
      - CIB guarantees of minimum user demand, fee rates, or total revenue, and fee-rate adjustments by a trusted third party comparable to a utility rate commission;
      - equity investment by the CIB that private investors view as partnership protecting project financial outcomes.
    - Match project profile with investor knowledge and needs using CIB staff market expertise and investor selection procedures.

- Brownfield investments (existing infrastructure) are attractive due to lower risks: development completed, known maintenance requirements and user demand. Recycling assets (sale of existing infrastructure) extracts capital for new investment but can require users to pay for longer periods or higher rates.

- The CIB’s role as filler of the funding gap:
  - CIB investment fills the difference between the total project funding required and the private investment determined by discounted FCF.
  - As discounted cash flows grow, the funding gap and thus CIB investment becomes smaller.
  - Actual FCF exceeding amounts needed to satisfy private investor returns should be distributed to the CIB.

- Additional points on CIB participation and investment structure:
  - Minimum private investment required for CIB to participate is not officially announced; lower private participation approximates PPPs.
  - To keep SPV majority-owned by private investors, any CIB investment greater than 49 percent equity would typically be loans; permissible forms of investment beyond 49 percent of total project cost should be determined.
  - Financial return allocation: FCF would first be allocated to private investors, and only then to the CIB for return of capital and financial returns. If FCF suffices for private-equivalent returns for the CIB, the project could be fully privately financed, implying the CIB effectively accepts lower returns to subsidize or be subordinate to private investors.
  - The estimated value of CIB risk mitigation efforts (including guarantees and non-financial staff contributions) should be reflected in total project cost regardless of recovery provisions in financing agreements; this raises valuation challenges.

- On investor rate-of-return discovery:
  - The risk-adjusted rate required to attract investors is not known precisely ex-ante; perceptions of risk vary with investor familiarity and capability.
  - Competitive bidding procedures should reveal investors’ true acceptable risk-adjusted rates (market rates).
  - Actual private returns during the project life can be controlled via allocation of FCF between private investors and the CIB.

### D. Effects on Public Finance and Strategic Objectives
- The draft CIB legislation’s objective of attracting private capital is general and provides limited guidance for CIB role, strategy, and operations; refinement is desirable to inform decisions on measurement and trade-offs.
- Issues to address for public finance implications:
  - How to measure maximization of infrastructure spending?
  - How the CIB spends allocated money?
  - How budget funds are allocated to the CIB?
  - Fiscal risks arising from the CIB and its investments.
  - Effect of the CIB on the efficiency of public investment.

- Possible objective refinements for attracting private investors:
  - Maximize total infrastructure spending by maximizing public and private financial resources.
  - Maximize total infrastructure spending under conditions of restrained public spending or borrowing dictated by fiscal policy.
  - Maximize infrastructure built by lowering project costs so the same spending purchases more infrastructure, regardless of financing source.

### E. Operational and Prudential Cautions Noted
- The CIB should be cautious when awarding guarantees in terms of number and potential payout to limit contingent liabilities.
- Valuation challenges exist in fully disclosing the value of CIB risk mitigation efforts and reflecting them in total project costs.
- The CIB’s investments and guarantees create explicit fiscal risks tied to the concessional nature of investments and to guarantees requiring Minister of Finance approval.

*Prepared by David Gentry (FAD) and Kotaro Ishi (WHD), excerpt from the IMF selected issues paper on Canada (June 21, 2017).*

### 16.      From a cost of funds perspective, private capital should be employed only if public

### 16.      From a cost of funds perspective, private capital should be employed only if public

### Cost of funds and choice between public borrowing and private capital
- The federal government of Canada currently can borrow using bonds yielding about 1.9 percent annually.
- Risk-adjusted rates of return required by private investors will likely be between 5 and 15 percent, assuming private investors bear at least some of the project risk.
- Given the above, the same project FCF could support more government borrowing than investor capital.
- Recommendation: To minimize financing costs, government should borrow to the maximum extent possible within a prudent fiscal framework before using private capital—unless public borrowing is constrained.
- Contextual constraint: The federal government wishes to limit borrowing and embark on a declining path of debt-GDP.
- Infrastructure demand context: Estimates of the infrastructure gap in Canada range widely from Can$150 billion to Can$ 1 trillion, implying government alone cannot meet demand.

### Tension between financial and social returns; role of user fees
- Financing infrastructure through user fees can reduce social returns if public infrastructure is used at less than its capacity.
- If public infrastructure has social benefits that exceed financial returns, and financial returns are insufficient to attract private investors, society is better off with public subsidies until capacity is reached; prices (user fees) should be used largely for demand management thereafter.
- Charging user fees to finance infrastructure may lead to unduly low demand and make society worse off.
- This rationale supports CIB funding at less than market rates of financial return.

### Potential offset from private sector expertise
- Private sector expertise could result in lower project costs or longer asset life, offsetting higher financing costs.
- The potential is enhanced if the CIB and private investor are joint equity investors, aligning interests to lower total costs cooperatively.
- Life cycle contracts with the SPV covering design, construction, operations, and maintenance would contribute to achieving lower total costs.
- Non-financial role for the CIB: document extent to which private implementation reduces total costs compared to similar government-implemented projects in Canada.

### Objectives for CIB deployment of taxpayer funds (implications for FCF and risk)
- Maximize the total amount of private capital leveraged
  - Advance by maximizing number of CIB projects and selecting projects with ease of implementation.
  - Encourage allowing private investors wide latitude to select attractive projects, resulting in investors focusing on least risky projects.
  - Effect: maximizes FCF, minimizes aggregate CIB investment, and contributes to relatively quick implementation.
- Minimize the risk-adjusted rate of return required by the private investor
  - CIB investment in a project is minimized.
  - Incentive for CIB to maximize non-financial methods for reducing investor risk (e.g., CIB staff handling early project development).
  - CIB would want wide latitude to choose projects from a relatively large pool.
- Advance projects where private sector management expertise can achieve significant economies
  - Favors: firms with specialization and experience; large, complex projects requiring sophisticated skills; projects benefiting from arm’s-length relationship from public pressures.

### Allocation and fiscal treatment of the CIB
- The CIB would be on-budget.
- Within the proposed Can$35 billion, the Minister of Finance will allocate funds as needed under the authority proposed in the CIB Act.
- Draft legislation is silent on treatment of return of CIB equity investments or retention of earnings.
- Draft legislation is silent on whether approval in the budget process of CIB investments is in the aggregate or on a project basis (affects CIB operating at arms-length).
- SPVs in which the CIB invests would likely not be on-budget.

### Fiscal risks and transparency issues
- Fiscal risks associated with the CIB and its investments are mainly implicit.
- CIB expenses include investment losses; such events lower government net worth but will not place an unexpected demand on spending because cash disbursements have already occurred.
- The CIB may issue guarantees which under the FAA are subject to approval by the Minister of Finance.
- The CIB will be protected from SPV liabilities by the corporate veil.
- Risks:
  - If an SPV is not required to publish its accounts, the federal government through the CIB may be implicitly liable to finish a project or operate infrastructure if the SPV fails.
  - Confidentiality issues may prevent the CIB from publishing the terms of its financing for the SPV.

### Efficiency implications under the PIMA framework
- Ways the CIB would enhance efficiency of public investment:
  - Lower project costs may result from private investor discipline in design, construction, and operation.
  - The CIB would need to publish project selection criteria and investor selection criteria as part of an open and competitive selection process.
- Ways the CIB might marginally reduce efficiency:
  - Assets initially owned and operated by the SPV may not be on a government asset registry; asset and asset depreciation likely not on government balance sheets, which could inhibit infrastructure planning.
  - If a separate planning process is established for the CIB, there is potential to fragment priorities; additional complexity arises if the CIB and private investors exercise discretion about project selection.

### Project and investor selection: principles and challenges
- Key issues to balance: public interest and priority; consistency with CIB objectives; promoting private expertise and innovation; resolving timing differences for project selection; selecting private investors competitively and transparently to establish market risk-adjusted rates of return and avoid favoritism.
- Project prioritization should be determined in government planning processes (municipalities, provinces, federal departments each have planning processes).
- Concern about proposals to establish a consolidated, federally-sponsored list of projects for the CIB: creating a new cross-jurisdictional planning process raises challenges in a federal system and contradicts good PFM practice that a single set of priorities should be established regardless of financing source.
- Advantages of allowing CIB and private investors latitude to select from a relatively large pool of high social-return projects:
  - Investor perceptions and expertise vary; a small project list could miss private participation opportunities and lead to higher demanded rates of return.
  - A larger pool allows investors to propose innovative solutions and allows the CIB latitude to identify projects that can be implemented more quickly with maximum leverage.
- Innovative solutions from private investors should be guided by government-stated priorities and typically focused on project delivery (design, construction, operations, maintenance).
- Unsolicited proposals for goals or projects not in the planning pool should not be entertained because they:
  - Undermine planning and priority-setting;
  - Invite inherently non-transparent bilateral discussions between investor and government;
  - Can be written to advantage the proposing investor, undermining competitive selection.
- If an investor proposes an alternative solution during the investor selection process, the proposal must be reviewed and consented to by the originating government before finalizing financing.
- Timing challenges:
  - Project FCFs depend on design capacity, user fees, and user demand; preliminary estimates may change with investor proposals.
  - The rate of return demanded by investors, and thus the proportion funded by private investors, cannot be known until the investor selection process is completed—this in turn determines CIB investment.
  - A two-stage process (short-listing followed by final selection on detailed proposals) might be employed to ensure fairness.

### Project selection strategies (implications)
- Strategy 1 — maximize infrastructure built:
  - i) maximize annual mobilization of private capital, including quantity and speed of project delivery;
  - ii) FCF is maximized (i.e. user fees are high and costs are reduced to the extent possible);
  - iii) the CIB takes on a relatively large amount of risk to lower risk-adjusted return to the investor and maximize leverage;
  - iv) a large pool of projects is available for investor selection;
  - v) investors will likely choose projects that are relatively simple and low risk.
- Strategy 2 — maximize social rate of return:
  - i) CIB equity investment at less than market rates of return is large;
  - ii) FCF is lower than potential (i.e. user fees are not maximized early when demand may be light) and thus the CIB is responsible for a higher proportion of total project funding;
  - iii) the CIB takes on relatively large amounts of risk;
  - iv) the planning process is relatively more important to identify projects of highest priority;
  - v) investor selection would adhere closely to planning priorities and requires a second review by the proposing government or planning process if proposed innovations may affect social rate of return.
- Strategy 3 — maximize marginal benefit of CIB funds:
  - i) the CIB leverages the most private capital per CIB dollar invested;
  - ii) the CIB emphasizes non-financial means to lower investor risk (e.g., CIB staff assisting early stage development);
  - iii) FCF is maximized (i.e. user fees are relatively high);
  - iv) the pool of projects should be large, allowing CIB latitude to find highest return projects;
  - v) the CIB will tend to encourage private investor innovation to lower upfront cost rather than long-term cost or effectiveness.

### Box 1 — Social versus financial returns (analytical summary)
- Framework:
  - Vertical axis: private rate of return.
  - Horizontal axis: social rate of return (includes externality effects).
  - Projects with a positive externality are on the right side of the 45-degree line.
  - Private sector will undertake projects if private rate of return exceeds cost of funding (Regions A, B, and F).
  - Government prefers projects in Regions A, B, and C; Region C has high social return but private return insufficient to cover costs.
- How the Infrastructure Bank (CIB) can help:
  - Government can provide financial incentives so that investor risk-adjusted rate of return increases (upward shift from X to X’).
  - Operational challenge: incentives should not be too generous, else private return after incentives could exceed social return (point X”).
  - Fair and transparent competition and professional expertise are required in selecting projects and negotiating financing terms.

### Comparative advantages of the CIB
- Two unique features of the proposed CIB among Canadian public infrastructure institutions:
  - Only the CIB is directed to invest in revenue-generating infrastructure (PPPs in Canada generally employ budget financing and PPP revenues typically deposited into the general treasury).
  - Only the CIB is directed to invest equity alongside private investors, signaling government commitment and establishing public-private partnership even with symbolic CIB investment.
- Equity partnership benefits:
  - Aligns interests of equity owners.
  - Enhances monitoring by providing access to SPV operational and financial information.

*Source: cr17211 - 16.      From a cost of funds perspective, private capital should be employed only if public (PDF chapter).*

### 33.      The issues identified in previous sections have implications for the CIB as an

### cr17211 - 33.      The issues identified in previous sections have implications for the CIB as an

### Implications for the CIB as an organization
- The CIB is better characterized as a fund than a bank. It cannot be financially self-sustaining. However, substantial benefits can be derived from the CIB as a fund.
- The respective roles of the CIB and the planning process must be clear:
  - The planning process is the primary channel for evaluating project social rates of return and determining a limited pool of projects available for private investment that enhance the public interest.
  - Asking the CIB also to evaluate social rates of return is redundant and could result in conflict with previously published priorities.
  - A clear role for the CIB would be to focus on financial issues integral to its strategy, and move projects previously screened for the public interest forward as quickly as possible.
- The skills set of CIB staff should reflect its role and strategy:
  - If the strategy emphasizes non-financial risk reduction, increases the intensity of interaction with governments, or focuses on social rates of return, more weight should be given to staff with government and public service skills.
  - If the strategy emphasizes creative financing, financial means of risk reduction, and matching of investor financial needs with project characteristics, more weight should be given to staff with capital markets skills.
  - This decision will likely affect total staff compensation.

### Possible public concerns
- The federal government may wish to address early in the public debate on the CIB objectives, policies, and procedures of concern to Parliament or the public. Explaining why certain policies and procedures are necessary to achieve widely-agreed objectives may make them more acceptable.
- Anticipated concerns include:
  - User fees:
    - Public surveys consistently report public resistance to user fees.
    - Revenue generated from infrastructure is perhaps the single most important distinguishing characteristics of the CIB.
    - User fees supplement government financial resources obtained through taxation, and increasing infrastructure investment requires additional revenue.
  - Perceptions of excess returns to investors:
    - The process for selecting investors must be open and competitive.
    - A competitive selection process must assure the public that private investor rates of return are the lowest possible rate given the risk of each project, and that a sub-set of investors do not receive any form of favoritism.
    - This process should parallel good procurement practices.
    - The CIB must be prepared to withdraw a project that requires unusually high risk-adjusted rates of return, and must have the staff capacity to make this assessment.
  - Subsidizing private investors:
    - CIB financing at a lower rate of return, writing-off investments, or granting funds may be necessary to give private investors the rates of return acceptable to them.
    - The public interest is served through such funding if the alternative is that government fully funds the project itself and bears all the project risk.
    - It is highly likely that many people will characterize such funding in support of private investors as subsidies, and as being unfair.
  - Input on project selection:
    - Concerns relate to the distribution of projects geographically and across jurisdictions, and how governments at all levels have a voice in the project selection process.
  - Input on project implementation:
    - All projects will be physically located in one or more provincial and municipal government jurisdictions, even if the project is under the responsibility of a federal department.
    - Local governments need to be assured of opportunities to provide input on the final design and implementation of the project, especially if innovative proposals have been made.

### Conclusions on the CIB
- The CIB is expected to be an effective new tool to increase investment in public infrastructure by mobilizing private capital. It adds to existing Canadian programs and resources in support of public infrastructure.
- The public interest would be served by instituting a process for selecting projects, to which private investment is invited, that builds on, without replacing, the existing infrastructure planning process at all levels of government.
- Private capital will be attracted by acceptable risk-adjusted rates of return. While private capital is costlier than government financing, private investors are expected to apply their expertise to lower total project costs, thus offsetting at least partially the higher financing costs.
- Private investors should be selected for specific projects in an open and competitive manner to minimize the risk of these rates of return being excessive.
- The CIB is expected to enhance public finances. User fees add to public funds available to support infrastructure. The size of private investment would be determined by the size of project-generated cash flows.
- The CIB will be on-budget in aggregate and will be reflected in the federal government balance sheet. Measures are available to contain explicit fiscal risks.
- The CIB should be thought of as a fund rather than a financially self-sustaining bank. Its main role is to match private investors with a short-list of high priority revenue-generating public projects, and find ways to advance those projects.
- The CIB will be unique in its ability to invest public funds in equity alongside private investors, thus aligning private capital and expertise with public interests.
- Communicating to the public CIB’s objectives, and policies and practices required to meet these objectives such as the need for user fees, is critical to gaining approval to establish the CIB and ensuring ongoing public support for its operations.
- Some public concerns can be anticipated and should be addressed proactively, such as resistance to user fees and public perceptions of high rates of returns benefiting private investors at government expense.

---

### ESTIMATING THE GROWTH EFFECT OF PUBLIC INFRASTRUCTURE: EVIDENCE FROM CANADIAN PROVINCES

### A. Introduction
- Canada has launched an ambitious infrastructure program to boost its long-term growth potential.
- The federal government is committed to investing Can$ 187 billion (equivalent to 7½ percent of GDP) in public capital, a half which (Can$ 96 billion) is new additional infrastructure spending. This will triple investment in public capital over the next decade.
- Priority projects include:
  - public transit infrastructure;
  - green infrastructure; and
  - social infrastructure (affordable housing, child cares spaces, and communities centers).
- Objective of the chapter: to estimate the macroeconomic effects of infrastructure investment using panel data for 10 Canadian provinces over 1961-2015.
- Key finding previewed: public infrastructure investment could boost economic growth, complement private investment, and generate positive spillover effects across regions: higher public investment in one province boosts output in the rest of Canada.

### B. The size of the problem
- Estimates of the infrastructure gap in Canada range from Can$150 billion to Can$ 1 trillion (Advisory Council on Economic Growth, 2016).
- World Economic Forum Global Competitiveness Index indicated Canada’s overall ranking in the quality of infrastructure dropped to 21st position most recently, down from 17th position a decade ago, with declines in roads and railroad rankings the most evident.
- The stock of economic infrastructure has declined:
  - from about 20 percent of GDP in 1961 to as low as 12 percent in 2007.
  - The downward trend was reversed during the 2008-09 global financial crisis when the government increased infrastructure spending as part of its stimulus efforts, but the stock of economic infrastructure assets remains at historic lows, 5¾ percent of GDP below the level in 1961.
- Economic infrastructure defined: railways, ports, airport, pipelines, roads, and energy networks that support delivery of public services, connect people to jobs, and facilitate businesses through reduction of costs of doing businesses.

### C. Policy considerations
- Canada’s fiscal federalism presents advantages and challenges in coordinating infrastructure investment:
  - The bulk of economic infrastructure assets is owned and managed by provincial and local authorities.
  - The federal government’s share of infrastructure assets has halved to 7 percent since 1990, while provincial and local governments have increased their share to more than 90 percent of total assets today.
- Advantages:
  - Lower levels of government are better positioned to identify local community needs; decentralization can be desirable for allocative and administrative efficiency.
- Challenges:
  - Many infrastructure facilities have connectivity and network effects; investing in one province likely has spillover effects to other provinces.
  - Currently, there is neither an established institutional mechanism for federal-provincial coordination nor a national infrastructure strategy that aligns priorities across all levels of government.
  - The federal government provides conditional grants to provincial and local governments to influence decision making over infrastructure projects, but their effectiveness is questioned.
- Recommended areas for reform (as suggested by the IMF Public Investment Management Assessment framework):
  - Specifying capital funding amounts by department in the medium-term budget framework, which could support more detailed sectoral investment strategies and a project pipeline.
  - Promoting inter-governmental coordination by consolidating federal and provincial approved projects into a single report updated annually.
  - Identifying major public investment projects in budget documents to clarify their contribution to achieving program objectives and protect their funding during budget execution.
  - Improving transparency regarding public investment spending by publishing Project Briefs and business cases for approved major projects, total expected financial obligations of ongoing projects, project audits, and PPP risks.
  - Establishing and publishing recommended methodologies for appraising proposed projects and evaluating projects upon completion.
  - Publishing the total cost of projects at the time of initial funding.

### D. Empirics — questions and approach
- Empirical questions:
  - Does infrastructure investment contribute to boost output? If so, how much?
  - Does infrastructure investment have spillover effects across Canadian provinces?
  - Does infrastructure investment crowd in private capital investment?
- Data and methods:
  - Panel regressions using data from 10 Canadian provinces, sample period 1961 to 2015 (annual data), balanced panel.
  - A standard Cobb-Douglas production function is estimated with infrastructure capital included:
    - Y_it = A L_it^{β_L} K_it^{β_K} G_it^{β_G} ḠO_it^{β_GO}
    - ḠO_it denotes real infrastructure stock in the other provinces.
  - Two weighting methods for ḠO_it:
    - Weighted by geographical distance (weights based on distance between provincial capital cities).
    - Simple average (overall infrastructure in all Canadian provinces).
  - Log-linear baseline model estimated:
    - y_it = β_i^L l_it + β_i^K k_it + β_i^G g_it + β_i^{GO} ḡo_it + u_it
    - u_it = α_i + γ_i ω_t + ε_it
  - Estimation methods:
    - Fixed effects model.
    - Mean group (MG) estimator (allows coefficients to vary across provinces).
    - Mean group with a common dynamic process (MG-CDP) to relax the constant time trend assumption.

### D. Empirics — preliminary empirical observations
- For the average of the full sample period, there is a relatively strong positive relationship between the annual increase in the stock of infrastructure assets and provincial GDP growth (Figure 3 shows a fitted line y = 0.86x + 1.2, R² = 0.5 in the scatter of average infrastructure capital growth rates vs average GDP growth rates for provinces over 1961-2015).
- The analysis follows literature showing public investment shocks have statistically significant and long-lasting effects on output (reference to IMF (2014) and Jordà (2005) local projection methodology), noting an unanticipated 1 percentage point of GDP increase in investment spending increased level of output by about 0.4 percent in the same year and by 1.5 percent four years after the shock (as context from cited literature).

*Italic: Source — cr17211 (PDF chapter/section) — content as provided.*

### 11.      The regression results are presented in Table 2.

### 11.      The regression results are presented in Table 2.

### Regression results — key findings
- In all models, the coefficients on private capital and employment are positive and highly significant.
- Fixed-effects models (Models 1–3):
  - Model 1: coefficient on economic infrastructure assets (G) is positive but not significant (0.0448).
  - Model 2: includes other provinces’ infrastructure assets (weighted by geographical distance). Own infrastructure and other provinces’ infrastructure coefficients are positive but not significant (G = 0.0463; GO, weighted average = 0.0470).
  - Model 3: includes the simple average of other provinces’ infrastructure assets. Own and other provinces’ coefficients are negative and insignificant (G = -0.0296; GO, simple average = -0.937).
- Mean Group (MG) estimators (Models 4–6):
  - Model 4: coefficient on own infrastructure positive but not significant (G = 0.0444).
  - Model 5 (simple average GO̅jt): coefficients on other provinces’ infrastructure are positive but not significant (GO, simple average = 0.0974); own infrastructure coefficients become negative and insignificant in some specifications.
  - Model 6 (weighted average GO̅jt): coefficients on other provinces’ infrastructure positive but not significant.
- MG-CDP estimators (Models 7–9):
  - Model 7: coefficient on own infrastructure assets is positive and significant (G = 0.0768*).
  - Model 8: including other provinces’ infrastructure assets (weighted), own infrastructure positive but not significant (G = 0.0626); other provinces’ infrastructure (weighted) negative and insignificant.
  - Model 9: performs relatively well — own infrastructure positive and significant at 15 percent (G = 0.105+), and other provinces’ infrastructure positive and highly significant (GO, simple average = 0.274***).
- Model-specific coefficients for Real private capital stock (K) and Employment (L) (selected values preserved exactly as reported):
  - Real private capital stock (K): Model 1 = 0.252***; Model 2 = 0.255***; Model 3 = 0.249***; Model 4 = 0.348***; Model 5 = 0.341***; Model 6 = 0.321***; Model 7 = 0.101***; Model 8 = 0.0836**; Model 9 = 0.104***.
  - Employment (L): Model 1 = 0.750***; Model 2 = 0.748***; Model 3 = 0.734***; Model 4 = 1.355***; Model 5 = 1.173***; Model 6 = 1.155***; Model 7 = 0.560***; Model 8 = 0.507***; Model 9 = 0.526***.
- Memorandum items:
  - Observations = 550 for all models.
  - R-squared (reported collectively) = 0.988 (for Fixed effects models 1–3).

### Interpretation and preferred specification
- There is some evidence that infrastructure assets are positively correlated with real GDP and have spillover effects across provinces.
- The estimates based on the MG-CDP model—which accounts for both a province-specific slope β and a province-specific time dummy γi ωt—are judged most plausible because the estimated coefficients are significant at reasonable levels.
- Models 7 and 9 suggest that the elasticity of infrastructure assets would be around 0.08 and 0.10, respectively.

### Marginal product of infrastructure assets (MPK)
- The MPK is calculated as the estimated elasticity of public infrastructure multiplied by the output to public infrastructure ratio.
- The MPK presented (Figure 4) for Canada is based on the estimated elasticity of 0.08 (Model 7).
- Time pattern and interpretation:
  - The MPK shows a steady increase between the early 1960s and mid-2000s, reflecting a decline in the stock of public infrastructure as a percent of GDP.
  - After the 2008-09 global financial crisis, as government investment in public infrastructure rose, the MPK started to fall.
  - The current level of the MPK remains above historical levels.
- Policy-relevant implication:
  - Combined with the fact that the current level of interest rates is much lower than its historical norm, this suggests there is room to further increase public infrastructure while keeping the return of infrastructure investment higher than the cost of funding.

### Crowding-in effects of infrastructure
- Model specification: infrastructure capital is assumed to affect real GDP through private capital via cross-product terms (g*k and go*k).
- Estimated cross-term coefficients (Table 3):
  - g*k (province i’s own infrastructure × private capital): 0.00839*** (Model 1).
  - go*k (other provinces’ infrastructure × province i’s private capital): 0.0134*** (Model 2).
- Interpretation: the results suggest evidence of crowding-in — public infrastructure is complementary to private capital, with highly significant positive interaction terms.

### Conclusion and policy implications
- Empirical findings:
  - Higher public infrastructure is correlated with higher economic growth.
  - Infrastructure investment can crowd in private investment.
  - Infrastructure investment has positive spillover effects across provinces.
- Policy recommendations implied by results:
  - More concerted efforts across different layers of jurisdictions to better coordinate infrastructure planning would maximize the economic benefit of infrastructure investment.
  - Given underinvestment of infrastructure over past decades and historically low cost of funding, investing in infrastructure would generate a net positive return.
- Elasticities and key quantitative pointers:
  - Elasticity of infrastructure assets ≈ 0.08 (Model 7) and ≈ 0.10 (Model 9).
  - Cross-term estimates indicating crowding-in: g*k = 0.00839***; go*k = 0.0134***.

*Source: cr17211 - 11.      The regression results are presented in Table 2. (PDF chapter/section).*

### 10.      Real estate taxation in both Australia and New Zealand tilts investment decisions

### 10.      Real estate taxation in both Australia and New Zealand tilts investment decisions

### Tax features that tilt investment toward housing
- Key elements that tilt investment decisions towards housing are tax concessions on capital gains and the ability to apply losses from negative housing investment gearing to income from other sources.
- New Zealand (2015): strengthened enforcement of existing capital gains taxation rules by introducing a presumption of resale intention at purchase if an investment property is being resold within two years (“bright-line test”).
  - A withholding tax for non-residents that resell within that period was also introduced.
- Australia (policy intentions and recent steps):
  - Intends to increase the withholding tax for non-residents in mid-2017.
  - Intends to exclude non-residents from the main residence capital gains tax exemption from mid-2019.
  - Introduced a penalty for non-residents who leave their property vacant for more than six months per year.
  - Within existing foreign investment approval rules for non-residents, foreign ownership in larger new apartment developments is planned to be restricted to 50 percent.

### Fiscal and regulatory measures targeting real estate investors
- As demand for real estate for investment purposes has been considerable, Australia and New Zealand put in place specific restrictions and measures targeting real estate investors.
  - Australia’s fiscal-based measures primarily target foreign investors.
  - New Zealand’s measures are aimed at speculative activity without differentiating between residents and non-residents.

- Australia: foreign investment approval and conditions (as described)
  - Real estate purchases by foreigners require foreign investment approval.
  - Only part of the approvals lead to actual real estate purchases, allowing for imperfect tracking of foreign buyers’ activity.
  - Non-resident foreigners generally need to apply for and receive foreign investment approval before purchasing any residential property in Australia.
  - Non-resident foreigners will normally be allowed to purchase new dwellings in Australia without being subject to any conditions.
  - There is no limit on the number of new dwellings a foreign non-resident may purchase, but approval is generally required prior to each acquisition.
  - Non-resident foreigners generally cannot purchase established dwellings as homes, for use as a holiday home or to rent out.
  - Foreign buyers have to provide citizenship and visa details, as well as Foreign Investment Review Board clearance, through the stamp duty process.

### State-level stamp duty surcharges in Australia (rates and timing)
- Victoria:
  - In 2015, additional stamp duty was introduced to any arrangement or transaction that involves the transfer of an interest in residential property to a foreign purchaser.
  - For transactions between July 2015 and July 2016, the surcharge was 3 percent.
  - After 1 July 2016, the additional duty rate is 7 percent.
  - There is also a 1.5 per cent surcharge on land tax.
- New South Wales:
  - In June 2016, a 4 percent stamp duty surcharge was imposed on foreign buyers of residential property.
  - The stamp duty surcharge was set to rise to 8 percent as of July 1, 2017.
  - In addition, the annual land tax on foreign homeowners would rise to 2 percent from 0.75 percent.
- Queensland:
  - In October 2016, 3 percent surcharge was levied on foreign purchases of houses and apartments.

### Effects of prudential and tax measures on prices, credit, and resilience
- Prudential measures in both Australia and New Zealand would appear to have had at least a temporarily moderating effect on house price increase and credit growth, and they helped to improve the resilience of household and bank balance sheets.
- Despite these measures, prices have still been increasing at a fast pace, further eroding affordability.

### Key takeaway (policy interaction)
- Coordinating prudential measures with tax-based measures is presented as a means to improve the overall effectiveness of macroprudential policy in cases when speculative and investment demand play a major role.

*International Monetary Fund — cr17211 - 10. Real estate taxation in both Australia and New Zealand tilts investment decisions*

### 2.      The discussion below is organized as follows. First, we analyze the effectiveness of

### 2.      The discussion below is organized as follows. First, we analyze the effectiveness of 

### A.   Assessing the Effectiveness of Macroprudential Measures in Canada

- Context and institutional feature:
  - More than a half of outstanding mortgage debt is covered by mortgage insurance due to a requirement that borrowers must obtain insurance for high loan-to-value (LTV) mortgages (with LTV over 80 percent).
  - Government-backed mortgage insurance is provided by the Canada Mortgage and Housing Corporation (CMHC) - a crown corporation with half the market share of new insured mortgages - and two private companies.
  - The government guarantees 100 percent of CMHC’s obligations and backs private mortgage insurers’ obligations subject to a deductible equal to 10 percent of the original principal amount of the mortgage loan.
  - To address risks associated with the provision of these guarantees, the government sets eligibility requirements for insured mortgages, serving as an important macroprudential tool alongside traditional regulators’ tools.

- Evolution of policy stance since 2000s:
  - Easing up to 2007:
    - Mortgage insurance rules were relaxed in the mid-2000s.
    - Together with lower interest rates, this boosted mortgage credit and housing prices and led to a sharp expansion of home equity credit lines.
  - Tightening since 2008:
    - Multiple rounds of measures tightened mortgage insurance rules and OSFI introduced numerous microprudential measures to strengthen bank balance sheets.

- Six rounds of macroprudential measures analyzed (2008, 2010, 2011, 2012, early-2016, late-2016) with additional premium increases in 2014 and 2015 noted but excluded from empirical exercise because changes were too small to impact credit demand.

- Methodology:
  - Two separate OLS equations estimated monthly over August 1998 to February 2017: one for mortgage credit year-on-year growth (Mt) and one for house price year-on-year growth (Ht).
  - Mortgage credit specification (1) includes matrix Xt of controls: unemployment rate, hourly wage growth, five-year mortgage interest rate, and house price growth; macroprudential measures are represented by dummy variables Dit for each round (2008, 2010, 2011, 2012, early-2016, late-2016).
  - Tested impacts at 1, 3, 6, and 9 months after implementation and over periods between rounds.
  - Also tested individual instruments by replacing Dit with changes in maximum LTV ratios and amortization period.
  - House price specification (2) includes lagged controls Yt: growth of completed but unsold houses, mortgage credit growth, nominal GDP growth, growth of sales of existing houses.
  - SUR and 3SLS estimators yield consistent results with OLS (note 5).

- Results on mortgage credit growth (estimated impacts by round):
  - 2008 round:
    - Estimated coefficient not statistically significant across specifications.
    - Credit growth decelerated in 12 months following measures, likely reflecting cyclical downturn, higher unemployment, and decline in household income rather than policy effect.
  - 2010 round:
    - Statistically significant impact, dampening mortgage credit growth by about 2.3 percentage points on average.
    - Package included lower LTV on refinancing loans, minimum down payment on non-owner-occupied properties raised from 5 percent to 20 percent, and tighter eligibility criteria.
  - 2011 round:
    - Reduced mortgage credit growth by about 0.9 percentage points on average.
    - Measures lowered LTV for refinancing and reduced maximum amortization from 35 to 30 years.
  - 2012 round:
    - Most effective: trimmed mortgage credit growth by an estimated 3 percentage points on average.
    - Measures focused on amortization, LTV for refinancing (to 80 percent), and DTIs; effect strengthened over time.
    - More than half of new insured refinance loans before 2012 had LTV higher than 85 percent; effective LTV for first-time buyers reduced from 100 to 95 percent.
  - Early 2016 round:
    - Did not seem to have an impact, possibly because it affected a relatively small proportion of the market (minimum down payment increase from 5 to 10 percent only for portion above $500,000).
  - Late 2016 round:
    - Seemed effective in reducing mortgage credit growth; more stringent qualifying rules had immediate effect by tightening underwriting for borrowers with down payment below 20 percent.
    - Results based on few observations and should be interpreted with caution.

- Results on individual instruments:
  - A one percentage point reduction in both the maximum LTV for new mortgages and for refinancing loans tends to reduce y/y credit growth by about 0.5 percentage points (Table A4).
  - Reducing the amortization period by one year appears to have lowered credit growth by 0.2 percentage points.
  - External evidence notes: CMHC (2011) suggested refinance volume dropped by 22 percent following 2011 measures; Dunning (2012) estimated 2011 measures pushed debt-service ratios above maximum for about 6 percent of high LTV mortgages taken in 2010 and about 11 percent of borrowers in 2011 would have been unable to access credit after amortization reduction.

- Link between mortgage credit and house prices:
  - Estimated house price equation indicates mortgage credit growth has a one-for-one effect on house price growth (Table A5).
  - Tighter mortgage insurance rules indirectly dampened house price growth by reducing mortgage credit growth.

- Counterfactual fitted results:
  - Without the measures, average mortgage credit growth since April 2010 would have been higher by more than 5 percentage points.
  - Household debt-to-income ratio would have been closer to 200 percent as of the third quarter of 2016, instead of the actual 167 percent.

### B.   International Experience

- Dataset and scope:
  - Updated the macroprudential dataset in Lim et al. (2013) covering 2000–16 across a sample of 35 economies.
  - Focus on four frequently used measures: limits on LTV ratios; caps on DSTI ratios; greater risk weights for banks’ credit assets; higher provisioning requirements for banks.
  - Panel regressions control for business cycle and cost of borrowing; GMM Arellano-Bond estimator used for house price and mortgage credit growth specifications with step-function variables for each instrument (increase by one when tightened, decrease by one when loosened).

- Main international findings:
  - All four measures can be effective in containing mortgage credit growth.
  - Economies that lowered caps on LTV and DSTI ratios, tightened provisioning requirements, or raised risk weights moderated mortgage credit growth on average by 0.9–2 percentage points (Table A6).
  - Lower caps on LTV and DSTI ratios have a statistically significant but economically small effect on house price growth, containing house price growth by about 0.4 percentage points (Table A7).

### C.   Conclusion

- Summary conclusion:
  - Canada’s macroprudential policy appears to have lowered mortgage credit growth, moderating the surge in house prices on a national basis.
  - The 2010, 2011, 2012 and late 2016 rounds of measures are estimated to have been effective, possibly due to their larger scope and size.

*Prepared by Ivo Krznar, Zsofia Arvai (both MCM), and Yulia Ustyugova (WHD).*

### 18.      International experience over 2000–16 suggests that tighter LTV limits and lower caps

### 18.      International experience over 2000–16 suggests that tighter LTV limits and lower caps on DSTI ratios could be effective in moderating both mortgage credit and house price growth.

### Key cross-country finding on macroprudential effectiveness
- International evidence (2000–16) indicates that tighter LTV limits and lower caps on DSTI ratios could be effective in moderating both mortgage credit and house price growth.
- In Panel GMM estimations (2000:1–2016:4, quarterly, seasonally adjusted):
  - For mortgage credit growth (real), LTV coefficient = -1.30*** (standard deviation 0.25).
  - For mortgage credit growth (real), DSTI coefficient = -0.87** (standard deviation 0.34).
  - For house price growth (real), LTV coefficient = -0.38*** (standard deviation 0.13).
  - For house price growth (real), DSTI coefficient = -0.44** (standard deviation 0.20).
- Other significant controls in panel estimations:
  - Mortgage credit growth (t-1) = 0.90*** (std 0.01).
  - GDP growth (t) in mortgage credit growth regressions ≈ 0.28–0.29*** (std 0.04).
  - Lending rates (t) in mortgage credit growth regressions ≈ -0.11 to -0.14*** (std 0.04).
  - House price growth (t-1) = 0.78*** (std 0.01).
  - GDP growth (t) in house price regressions ≈ 0.34–0.36*** (std 0.03).
  - Lending rates (t) in house price regressions ≈ -0.48 to -0.55*** (std 0.06).
- Sample sizes:
  - Mortgage credit panel: N = 2117 observations, Countries = 35.
  - House price panel: N = 1406 observations, Countries = 30.

### Canada-specific context and policy stance
- Canada’s DSTI limits are described as relatively stringent compared to other economies in the sample.
- Canada’s limits on the LTV ratio are above the average in the sample of economies.
- The assessment: Further tightening of the LTV ratios could help contain growth in mortgage credit and house prices.

### Empirical results for Canada (monthly/quarterly estimations covering 1999:2–2017:2 and 1999:2–2017:2 where noted)
- Mortgage credit equation (dependent variable: Mortgage credit (y/y); estimation period 1999:2–2017:2; monthly, seasonally adjusted):
  - House prices coefficient = 0.08** (std 0.03).
  - Wages coefficient = 0.35*** (std 0.10).
  - Interest rate coefficient = -1.84*** (std 0.17).
  - Unemployment rate coefficient = -1.88*** (std 0.20).
  - 2008 measures coefficient = -0.56 (std 0.49).
  - 2010 measures coefficient = -2.26*** (std 0.53).
  - 2011 measures coefficient = -0.94** (std 0.45).
  - 2012 measures coefficient = -3.07*** (std 0.29).
  - 2016 measures 1 coefficient = 0.31 (std 0.42).
  - 2016 measures 2 coefficient = -1.33*** (std 0.38).
  - Number of observations = 217.
  - R^2 = 0.84.
- Effectiveness of measures at different horizons (Mortgage credit equation, regressions I–V):
  - 2008 measures: 1 month = 1.09; 3 months = 1.51; 6 months = 1.34; 9 months = 1.11; Between rounds = -0.45. Corresponding standard deviations: 0.81, 0.99, 0.91, 0.88, 0.61.
  - 2010 measures: -2.43***, -2.69***, -2.38***, -2.55***, -2.58***. Standard deviations: 0.81, 0.81, 0.70, 0.61, 0.56.
  - 2011 measures: -0.64, -0.47, -0.38, -0.69, -1.04**. Standard deviations: 0.51, 0.42, 0.40, 0.46, 0.47.
  - 2012 measures: -1.60***, -2.01***, -2.10***, -2.19***, -3.06***. Standard deviations: 0.26, 0.37, 0.32, 0.31, 0.31.
  - 2016 measures 1: 0.73**, 0.74***, 0.59, 0.26, 0.26. Standard deviations: 0.27, 0.27, 0.35, 0.43, 0.43.
  - Note: *,**,*** indicate significance at the 10, 5, and 1 percent levels respectively. Newey-West consistent variance estimator used; regressions I–IV estimate effects after 1, 3, 6 and 9 months; regression V estimates effects between rounds.
- Effects of specific measures on Canadian mortgage growth (OLS, dependent variable: Mortgage credit (y/y); 1999:2–2017:2; monthly):
  - Amortization period coefficient = 0.19*** (std 0.02). Number of observations = 217. R^2 = 0.59.
  - LTV on refinance loans coefficient = 0.53*** (std 0.04). Number of observations = 217. R^2 = 0.74.
  - LTV on first time borrowers coefficient = 0.55*** (std 0.10). Number of observations = 217. R^2 = 0.54.
- House price equation for Canada (OLS, dependent variable: House prices (y/y); 1999:2–2017:2; monthly):
  - Mortgage credit coefficient = 0.99*** (std 0.18). Number of observations = 217. R^2 = 0.48.
  - Completed houses coefficient = -0.11*** (std 0.04).
  - Existing sales coefficient = 0.10*** (std 0.03).
  - GDP coefficient = 0.30* (std 0.20).

### Micro- and macroprudential policy timeline highlights for Canada (selected measures and dates)
- June 2011 (effective January 1, 2013): Protection of Residential Mortgage Hypothecary Insurance Act (PRMHIA) and amendments to the National Housing Act assented.
- March 2012: Economic Action Plan 2012 announcements (prohibiting covered bonds backed by government-insured mortgages; CMHC designated administrator of covered bond framework; CMHC mandate enhanced to include financial stability; CMHC commercial activities subject to OSFI examination).
- June 2012: Guideline on Sound Residential Mortgage Underwriting Practices (B-20); Maximum LTV on the revolving portion of HELOCs cut (from 80 to 65 percent); Stated Income mortgages no longer allowed without some verification of income.
- February 2014: Economic Action Plan 2014: guarantee fees 3.25 percent of premiums written and 10 basis points on new portfolio insurance written; CMHC reduced annual limit of portfolio insurance issuance from $11 billion to $9 billion; authorized NHA MBS and CMB limits for 2014 set at $80 billion and $40 billion respectively.
- September 2014 (effective January 1, 2015): Revised Minimum Capital Test Guideline for property and casualty insurers.
- May 15, 2015: Amendments to PRMHIA prohibiting substitution of loans in portfolio insurance pools (full implementation June 30, 2015).
- Residential Mortgage Insurance Underwriting Practices and Procedures (Guideline B21) issued (announced November 2014, full implementation June 30, 2015).
- December 11, 2015: CMHC announced changes to securitization programs; guarantee fee schedule changes effective July 1, 2016; for 2016, Minister authorized $105 billion for NHA MBS and $40 billion for CMB.
- December 11, 2015 (effective November, 2016): OSFI updated regulatory capital requirements for residential mortgages (risk-sensitive floor tied to increases in local property prices and/or house prices high relative to borrower incomes; updated standardized approach for federally regulated private mortgage insurers).
- February 10, 2016 (effective July 1, 2016): Department of Finance required portfolio-insured loans be funded only through CMHC securitization programs.
- October 3, 2016: Consultation launched on lender risk sharing to limit federal government’s financial obligations in mortgage defaults and promote prudent lending.

*Source: cr17211 - 18.      International experience over 2000–16 suggests that tighter LTV limits and lower caps on DSTI ratios could be effective in moderating both mortgage credit and house price growth.*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2017/cr17211.pdf_