Is Defence Spending on Target or is the Federal Government Shooting Blanks?

by Alex Reeves and Randall Bartlett

The old adage, “The more things change, the more they stay the same”, seems to be the cornerstone of the Government of Canada’s approach to defence policy. The federal budget was released in late February, and left many questions relating to the Department of National Defence (DND) unanswered. Specifically, where is the money that was promised in the newest Defence Policy Review (DPR)? In the budget, the federal government mentioned defence only a handful of times and didn’t address much of the additional funding that was promised. Now, with the publication of the DND’s Departmental Plans (DPs) – the federal government’s 3-year departmental spending plans – we can better see the federal government’s short-term vision.

While the DPR promised “stable, predictable, realistic funding”, the federal government’s plans seem to have come up short. To illustrate the disconnect between the Government of Canada’s plans and their promises, we use multiple documents to compare the impact of separate defence funding outlooks. By comparing the 3-year planned spending from DND’s most recent DP and the long-term forecast provided in the DPR, we can underline the differences between what the government promised in the DPR and what their current plans seem to be. In addition, we will compare different fiscal documents to show historic trends when compared to the DP estimates, and their implications for defence spending on different accounting bases.

It should be noted that some research has been done to explain the impacts of not spending earmarked defence funds have on government finances. For instance, an article published by CBC in 2015 discussed the longer-term historic decline in defence spending as a share of Gross Domestic Product (GDP), and the impacts that both budget cuts and spending lapses have had on the deficit. But, while this is a fruitful area for future research, it won’t be discussed at length in this paper.

Hey Big Spender

It was roughly a year ago that the Government of Canada first published the most ambitious military funding plan in the last decade, promising billions of additional dollars to fix much-needed equipment and reinforce weakened regiments. To show that the federal government was going to fulfil this promise, the DPR went into a great detail on how the military would receive its funding going forward. Table 1, taken from the DPR, lays out DND’s planned spending for the next 20 years.

The Big Shortfall in Promised Defence Spending

So, how do the promises made in the DPR measure up to the newest spending plans seen by Parliament in the DPs? With mixed results. The DP 2018 shows considerable differences to both the DP 2017 and the DPR. To its credit, DP 2018 is showing similar expenditure numbers to the DPR in the 2017-18 and 2019-20 fiscal years on a cash basis. Where the DP falls short is in the 2018-19 and 2020-21 planned spending. Chart 1 shows a comparison of forecasted spending from DP 2017 and DP 2018, as well as the DPR. Over the four years of data available in the newest DP, total net spending is forecast to be $5.7 billion lower than the spending outlined in the DPR, most of which stems from a $4.3 billion difference in the 2020-21 fiscal year.

What would the implications be if the federal government does come up short in funding military spending by $4.3 billion in fiscal 2020-21?

Looking at it from a cash basis, if the money were to be held back exclusively from capital spending, it would impact the procurement and upkeep of major equipment such as planes, ships, and tanks. Using the DPR funding forecasts as a benchmark, and given the DPR estimated major-equipment-related spending to be 21.83% of the $24.3 billion in defence spending in the 2020-21 fiscal year, capital spending would equal $5.3 billion in fiscal 2020-21. But if the DP 2018 spending forecast for fiscal 2020-21 of $20.1 billion is used instead while maintaining the 21.83% share of the overall budget for capital expenses, capital spending would be closer to $4.3 billion. In either of these scenarios, a disappearance of $4.3 billion from planned spending in fiscal 2020-21 (see Chart 1) would be equivalent to an 80% or 100% decrease in funding for major equipment purchases in that year.

As noted in the paper “Strong, Secure, Engaged So Far” by Dr. David Perry, the capital funding promises made in the DPR seem to be out of reach. In his paper, Dr. Perry outlines the promises made in the DPR, comparing its capital forecasts to the historical spending that occurred during the ramp up and withdrawal of major operations in Afghanistan. He concluded that “If DND can achieve in peacetime the same rate of spending increase reached a decade ago while the country was at war, it will spend significantly less than projected on capital under SSE.”  This reinforces the possibility that the forecasted difference in spending between the DP 2018 and the DPR could be related to a substantial shortfall in anticipated capital spending.

In an operating expense context, each of the approximately 90,970 defence-related full-time equivalents (FTEs) in 2018 is estimated to cost an average $99,980 per year for salary and benefits. With that in mind, the $4.3 billion in reduced spending would be equal to 46% of the 2018 expenditure estimates for personnel, or equal to roughly 42,200 FTEs. Even if one were to adjust for future growth in personnel and compensation, this number of FTEs would be only slightly lower. Keep in mind that spending on other operating expenses, such as repairs & maintenance and utilities, are also correlated to the number of FTEs. This means the $4.3 billion lower spending number would likely be reached with a smaller reduction in FTEs than the 42,200 estimated above.

While it is impossible to see the future, it is easy to see that defence spending below committed levels would likely come from multiple areas as opposed to specifically operating or capital spending. But these scenarios at least serve to illustrate that the shortfall in planned spending will be substantial, if realized.

Accounting for Differences in Accounting

How does the DPR fair when looking at the budget from an accrual accounting point of view in addition to a cash basis? Unlike any other government spending documents, Table 1 from the DPR shows long-term future defence spending on both an accrual- and cash-accounting basis. This isn’t something the federal government does often, usually leaving the differences between cash and accrual as an almost undecipherable riddle. As such, the federal government should be applauded for taking this step toward greater transparency and accountability.

In the case of possible reduced spending on capital investments, it would make little difference to the accrual numbers because capital expenditures have their costs spread over the useful life of the assets as opposed to being booked in the year it is paid for. For example, if the useful life of a new fighter jet is 30 years, the federal government books it as a small expenditure each year over the 30-year period on an accrual accounting approach. In contrast, it would be booked as a single outlay when using cash accounting. However, in the operating expense scenario, a sharp decline in spending in a single year would heavily impact the accrual forecast. This is because much of the cost of operating expenses, such as salaries, are booked in the year they are spent, and not spread over time.

But are the accrual forecasts provided in the DPR accurate when compared to existing government documents? It doesn’t appear so, as there are major inconsistencies when using the DPR as a benchmark for accrual funding relative to other federal government fiscal publications. For instance, in fiscal 2016-17, the DPR reported a $17.1 billion accrual expenditure line. This was contrasted by the Public Accounts of Canada, which put accrual expenditures by DND at closer to $25.5 billion in the same year – a whopping $8.4 billion difference.  

Major differences in reported numbers for DND on a cash and accrual basis go back even further if you compare historical spending from the Treasury Board Secretariat (TBS) with the Public Accounts. As seen in Chart 2, between the 2012-13 and 2016-17 fiscal years, there was a range of annual differences in cash and accrual spending from $2.7 billion to $9.8 billion dollars, averaging $5.5 billion per year over the five-year period. The differences in the cash and accrual spending estimates were all booked in the Public Accounts under “Accrual and other adjustments”. This means that these increases in accrual adjustments could be anything from capital amortization to changes in pension obligations.

These differences are further underlined when incorporating the defence spending data published by North Atlantic Treaty Organization (NATO) each year. While these numbers are expected to be marginally different due to differences in how defence spending is defined, e.g. NATO intentionally excludes pension obligations, they remain far more in line with the accrued numbers published in the Public Accounts than to the forecasted numbers presented in the DPR.

The most likely suspect for the increasingly large divergence in the cash and accrual numbers is a mix of rising pension obligations and capital spending. In the case of pension obligations, the numbers reported in the Public Accounts are likely going up, in part, due to the inverse relationship between interest rates and long-term pension liabilities. Over this historical period, the decrease in interest rates would have increased the amount the Government of Canada needed to spend each year, on an accrual basis, to continue meeting its pension obligations. This is discussed in more depth in work done by the Institute of Fiscal Studies and Democracy (IFSD), as well as Parliamentary Budget Office. In addition, every three years there is a review of the military pension structure, which have coincided with previous expansions in the differences between the cash and accrual estimates of defence spending. Going forward, the IFSD anticipates that rising interest rates will lower the future pension obligations of the military, and public service more broadly.

Regarding capital amortization, changes in accrual adjustments could be caused by failures to successfully procure new equipment, overspending on equipment, and a lack of information. A 2013 Internal Audit of Capital Project Cost Estimation from DND found that, “Insufficient rigour was applied to the cost estimation assumptions in the project identification phase due to insufficient standards for original cost estimates.” The report went on to say that, “In subsequent project phases, schedule delays, scope increases, unpredictable inflation and delays in engaging industry have contributed to cost growth.” These higher procurement costs could have contributed to a small degree of the changes present in the Public Accounts’ accruals. In addition, a 2016 report from the Office of the Auditor General (OAG) on operating and maintenance support from military equipment concluded that, “National Defence had made some initial planning assumptions that overestimated equipment use, underestimated support costs, and under-resourced personnel requirements. These assumptions led to higher costs and reduced equipment availability for training and operations.” The OAG report also goes on to explain how delays in procurement, such as the decade wait for Cyclone Helicopters, could have increased accrual costs by having drastically increased the lifespan of the equipment due to maintenance and repair costs.

The reality of the accrual adjustments is that all the discussed factors would contribute to a degree of changing estimates, pushing up the long-term costs of DND. The bigger question is: Why the documents being published, such as the DPR and TBS spending reports, seem to be calculating their accrual adjustments using a different approach than those provided by the Public Accounts and NATO?

The 2% of GDP NATO guideline might be a bridge too far

All in all, the future of Canadian defence spending is unclear, and the promises to revolutionize it may or may not come to fruition. In Table 2 of the DPR (below), the Government of Canada promises defence spending (on a cash basis) will rise beyond 1.4% of GDP by the 2024-25 fiscal year. The DPR argues that while this is far below the 2% guideline published by NATO, it is sufficient to ensure Canadian operational capabilities and support for both our national and international obligations.

To examine this claim, we divided the defence spending forecast published in Table 1 of the DPR (going out to 2026) by the nominal GDP forecast published by the federal Department of Finance in their Fall Economic Statement 2017 (FES 2017). It should be noted that we grew out the level of nominal GDP published in the FES 2017 by the growth rates published in the Department of Finance’s December 2017 Update of Long-Term Economic and Fiscal Projections 2017. It should also be noted that Chart 3 spending estimates are reflective of additional funding allocated in FES 2017.

As seen in Chart 3, the cash projections provided in Table 2 fail to meet the 1.4% of GDP level by the 2024-25 fiscal year. Even if you were to boost the cash numbers in 2024 by the 0.17% contributed by other departments, this forecast would be at 1.29%, 0.11% lower than the 1.4% promise. Based on the numbers collected from the TBS, Department of Finance, and DND, it appears that the only time defence spending could have peaked past 1.4% in the last 20 years is in the 2016-17 fiscal year, and only when using the accrual numbers provided in the Public Accounts.


Like Pandora’s Box, the further one delves into the black box of government funding for national defence, the worse things start to look and the fewer questions get answered while more arise. With all the inconsistencies between planning and reporting documents as pertains to military spending, and both internal and external reports critiquing the Government of Canada’s track record, the only hope is that one day the federal government will become as transparent as their rhetoric suggests. Indeed, this would go further in supporting the military than big spending announcements that consistently fall short.