From the 1950s to the 1970s, borrowing was one of the single largest sources of municipal infrastructure financing. But after the deficit and debt scare of the 1990s, and the fiscal belt-tightening that followed, borrowing on the public credit became almost universally despised. The conventional wisdom that developed is that all government expenditure—including infrastructure—should be met out of current revenue. No borrowing. Period. While the point may have some merit in the federal and provincial context, it certainly doesn’t in the municipal context.
Municipal Budgets are Unique
First, municipal budgets are much more capital intensive. In 2010, the City of Calgary’s infrastructure investment was 54% of its operating expenditure. Capital expenditures and grants in the 2011 Alberta budget are only 19% of operating expenditure.
Second, municipalities typically face a few very large projects. Federal and provincial capital budgets are filled with dozens of relatively small projects. Unlike federal and provincial governments, municipalities have very little flexibility to finance their capital needs by timing infrastructure projects.
Third, municipal borrowing is entirely different than federal or provincial borrowing. Most federal and provincial debt has been acquired as a result of operating deficits—borrowing to cover ongoing expenses like the provincial payroll. By law, municipalities in most provinces cannot run operating deficits. When municipalities borrow, it is always for infrastructure. The difference here is between “smart” debt and “stupid” debt. “Smart” debt equates to the mortgage on a home, where the debt incurred is offset by a valuable capital asset. “Stupid” debt is incurred to consume, like buying groceries on a credit card and then carrying the balance month after month or even year after year.
Fourth, the conventional wisdom is excessively conservative. Corporations with a strong balance sheet borrow. Why? Because it makes sense to grow the business by financing productive assets with debt. The same applies to local governments and infrastructure. The absence of any tax-supported debt is not the litmus test for fiscal responsibility. Rather, fiscal responsibility involves balancing the operating budget over the business cycle and maintaining or increasing financial net worth across the long-term. None of this is an argument against borrowing for infrastructure.
In short, a completely debt-free city should never be the ultimate goal of fiscal policy, regardless of how well it plays politically. This is especially the case if the trade-off is an underfunded stock of capital assets. The “pay-as-you-go” approach is arguably better for a city fiscally, but it does not always contribute to the overall health of a city, which certainly encompasses more than the balance sheet.
The concept of “smart” debt has emerged to help bolster support for borrowing as a valid form of infrastructure financing. The concept seeks to build consensus around the use of debt by emphasizing its role as part of any long-term capital plan, and recognizing that “pay-as-you-go” cannot accommodate all infrastructure needs. “Smart” debt sets out broad parameters on how a city should borrow. The idea comprises five elements.
1) Appropriate projects: Ideal infrastructure for borrowing is large and expensive, has a long lifespan, is one-time or non-recurring in nature, and where borrowing can lever additional financing elsewhere.
2) Debt levels: Smart debt means sustainable borrowing by using some notion of “optimal” debt relative to current operating revenues and anticipated growth of that revenue. Smart debt requires cities to work through the thorny question of their tolerance for debt. In February of 2002 for example, the City of Calgary implemented a new capital financing policy that allowed for significant new borrowing. But, strict limits were set. The cost of servicing all tax-supported debt could not exceed 10% of tax-supported expenditures. In October of 2002, the City of Edmonton also approved a new debt policy. Total debt charges were not to exceed 10% of city revenues and debt charges for tax-supported debt were capped at 6.5% of the tax levy.
3) Amortization: Smart debt does not see amortization terms set arbitrarily, or with the sole consideration being lowest cost. Rather, amortization terms reflect the life of the asset. Amortization terms in Canada today tend to be in the 10-20 year range, but in the past, 25-30 year debentures were not uncommon, and they are still in use across the US. Longer amortization lowers the costs of debt servicing and also allows more borrowing to occur. While this does mean paying more interest, a lot of that interest is offset by avoiding the costs of future inflation. Longer amortization is more than reasonable for assets with a life span of 50 or even 100 years. A good example comes from the City of St. John, New Brunswick, where the City proposed a significant user fee increase to fund a 20 year debenture for water and sewer improvements. The local Chamber of Commerce, balking at the magnitude of the rate increase, proposed that the debt term be extended to 30 years, the maximum allowed under provincial legislation. The Chamber argued that the user fee increase could be almost halved.
4) Debt structure and technique: In Canada, most municipal borrowing occurs through amortized debenture bonds where both interest and principal are paid in equal installments. But, the world is full of other options. One example is “retractable” or “bullet-style” debt where only the interest is paid for the first half of the term with principal payments coming on line for the second half. This debt can be used to advance desperately needed infrastructure. Smart debt argues for using a variety of borrowing structures and techniques:
- Municipal Tax-free Bonds
- TIF Bonds
- Local Improvement Bonds or Special Assessments
- Revenue Anticipation Notes and Revenue Bonds
- Bond Banks
- Revolving Loan Funds
- Senior Government Infrastructure Banks or Credit Enhancements
5) Repayment: Smart debt recognizes that borrowing can only finance infrastructure—the borrowing itself must be funded. Before issuing debt, cities draw up a comprehensive repayment plan. A good repayment plan incorporates the concept of earmarked taxation to build public support for increased capital spending and the issuance of debt. It’s easier to sell the public on incremental tax increases when they are earmarked for projects that are highly valued. The City of Calgary has earmarked special property tax levies to fund borrowings, and so has the City of Edmonton. The City of Saskatoon also expressed interest in earmarking a portion of the federal fuel tax revenue to repay debentures.
How Much Debt?
The most contentious feature of any “smart” debt policy is building a consensus around what constitutes a tolerable level of borrowing. Achieving agreement here is difficult because of the subjective nature of the question. At the same time, there is a way to conceptualize the issue and sharpen the focus. The process starts by recognizing that the tax revenue of most governments tends to grow over time. Against this tax revenue growth several “scenarios” can be plotted (Figure 1).
When debt—or the cost of servicing debt—is growing faster than tax revenue, the trend is both unreasonable and unsustainable. This is Scenario #1. The growing cost of debt will continue to chip away at tax revenue and crowd out other expenditures. At the opposite extreme is a steadily decreasing level of tax-supported debt, which given the huge infrastructure funding challenge, is just as unreasonable. This is Scenario #4. A reasonable and sustainable level of debt lies somewhere between Scenarios #2 and #3, which would see the outstanding stock of debt and debt servicing costs increasing over time, but never at a pace that outstrips tax revenue growth. There is no deterioration in a city’s fiscal position if outstanding debt and the costs of debt servicing grows in proportion to the growth in revenues—assuming of course that taxes are not intentionally raised beyond reasonable levels.
The point is that “runaway” debt and a “debt-free” city are both extremes to be avoided. Between the two lies a reasonable and sustainable level of debt.
Debt is Becoming “Smarter” in the West
Western Canada’s large cities have not always fared that well considering the general concept of “smart” debt. The Edmonton experience is illustrative (Figure 2). Starting in 1990, Edmonton embarked on a quest to decrease its debt. By 2003, tax-supported debt had fallen from $200 million to $24 million. Meanwhile, tax revenues continued to grow, as did the city’s infrastructure funding gap. Throughout most of the 1990s, the City of Edmonton was arguably following an “unreasonable” debt policy. Since 2000, however, tax-supported debt has grown, which reflects a change in borrowing policy.
Edmonton is not alone (Figure 3). This is the same pattern reflected across most of the West’s big cities. The City of Regina actually succeeded in eliminating all tax-supported debt, and Saskatoon came very close. Only the City of Vancouver has seen debt levels growing in tandem with tax revenues. However, recent increases in tax-supported debt do represent a much more balanced view of the role that borrowing can—and should—play in infrastructure financing. This is not a situation to be bemoaned, but applauded.
One of the reasons is that the timing for borrowing just could not get any better. Interest rates today are at the lowest point seen over the past 50 years (Figure 4). For cities with the capacity and the need to borrow, there may be no better or cheaper time than now.
To round out discussion over the smart debt option, it is important to understand the three stages of addressing a deficit—whether it be a budget deficit or an infrastructure funding deficit. First, growth in the deficit needs to be arrested (ensure the bleeding does not get worse). Second, the deficit needs to be closed (the bleeding must be staunched). Third, the accumulated infrastructure “debt” resulting from annual “deficits” needs to be addressed (the spilled blood needs to be cleaned up).
The potential of smart debt operates within the first step. There are four different approaches. The first approach (Figure 5) sees the entire annual funding gap (the red line growing over time) financed in the short-term by debt. Here, debt solves the short-term funding crunch but the amount of debt quickly bumps up against a previously set tolerance level. At that point, borrowing must essentially stop. The funding “gap” reappears, and continues to grow. Little has been gained.
The second approach (Figure 6) sees borrowing ramping up over the short-term after which the pace slows to keep debt levels tolerable. This addresses immediate high priority needs, but may not arrest long-term growth in the funding gap.
The third approach (Figure 7) sees modest borrowing annually against an operating budget that is growing as well. If borrowing proceeds at a slightly slower pace than the growth in operating revenues, then the costs of servicing debt relative to the budget do not rise and debt can be used more effectively over time. This may have the potential to arrest at least some of the growth in the infrastructure funding gap over a longer-term.
A variation on this approach (Figure 8) is to borrow substantially, but only in certain years. In the intervening years, debt is repaid, but then ramped up once again. Over the years, this has tended to be the general approach taken by the City of Saskatoon.
Getting “Smarter” Yet
From a fiscal policy perspective, it is good to see some “pick-up” on the smart debt concept. But more could be done. One example would be for Canadian cities to secure a wider range of borrowing tools.
At the Canadian Construction Association’s AGM in March of this year, I had a unique opportunity to spend some time in conversation with a former executive of Bird Construction. He offered up the idea of establishing a federal infrastructure bank or revolving loan fund. This would be a pool of capital created by the federal government using its AAA + bond rating to secure funds at the lowest possible rates of interest and make that available to municipalities. In his view, the idea represented the type of national commitment needed to address the infrastructure challenge.
When it comes to borrowing, there just could be no better time. The Bank of Canada rate averaged 0.65% in 2009, 0.85% across 2010, and is now sitting at 1.25%. It’s the cheapest money available since at least the 1960s, and cities would be well advised to use that to their advantage.
– By Casey Vander Ploeg, Senior Policy Analyst