Skip to content
Module 01 of 845 min readAdvanced

Why governments borrow

Deficits, tax-smoothing, and the intertemporal budget constraint — the legitimate case for debt before the trouble starts.

13%

Listen along

Read “Why governments borrow” aloud

Plays in your browser using on-device text-to-speech — nothing leaves the page.

Learning objectives

By the end of this module, you should be able to:

  • 01Explain the legitimate reasons governments borrow
  • 02State the tax-smoothing rationale and the intertemporal budget constraint
  • 03Distinguish good borrowing (investment) from bad (consumption)
  • 04Recognise where Ricardian equivalence and deficit bias bound the case

Public debt is neither inherently good nor bad — it is a tool, and like any tool it can build or destroy depending on use. This course is about that tool: how it works, when it becomes dangerous, and what happens when it breaks. We begin with the legitimate case for borrowing, because you cannot judge when debt is excessive without first understanding when it is wise.

Why borrow?

  1. Tax-smoothing — finance temporary spikes in spending (a war, a drought, a pandemic) by borrowing, so that tax rates can stay stable rather than spiking and crashing. Volatile tax rates are distorting (the deadweight loss rises with the square of the rate, the Public Finance course), so smoothing them with debt is efficient.
  2. Financing investment — borrow to build infrastructure that yields returns over decades, matching the financing to the life of the asset and to the future generations who will benefit (intergenerational fairness — those who use the road help pay for it).
  3. Counter-cyclical stabilisation — run deficits in downturns to support demand (the legitimate Keynesian case, bounded by the deficit-bias caveat below).
  4. Bridging timing mismatches — smooth the gap between lumpy revenue and continuous spending within the year (the cash-management function of the Budgeting course).

Tax-smoothing, formally

Barro's tax-smoothing

Robert Barro (1979) gave the canonical theory: a government should hold tax rates roughly constant over time and use debt to absorb temporary fluctuations in spending and revenue. Because the efficiency cost of taxation rises more than proportionally with the tax rate, smoothing the rate across good and bad years minimises the total distortion. So a temporary surge in spending should be debt-financed (spread the cost smoothly over future taxes) rather than met by a sharp, distorting tax hike. This is the cleanest economic justification for deficits — and note what it implies: the deficit should be temporary and reversed, with debt rising in bad times and falling in good. Permanent deficits are not tax-smoothing; they are something else (the deficit bias of the Public Choice course).

The intertemporal budget constraint

Debt is a claim on future surpluses

Borrowing is not free money; it is a claim on the future. The government's intertemporal budget constraint states that the present value of future primary surpluses must equal the current stock of debt: debt today = PV(future primary surpluses) In plain terms: every shilling borrowed must eventually be matched by a shilling of future revenue over spending (a primary surplus), discounted to today. Debt does not have to be repaid in full at any date (it can be rolled over), but the path must be consistent with generating enough future surpluses to service it. A government that borrows without any credible future surpluses to back the debt is on an unsustainable path — the arithmetic of the next module. The constraint is what separates borrowing (a claim to be honoured) from a Ponzi scheme (rolling debt forever with no backing), which markets eventually refuse to finance.

Good debt and bad debt

The use of the borrowed money matters enormously. Borrowing to finance productive investment that yields a return above the cost of borrowing is self-justifying — the asset (a port, a power plant, an educated population) generates the future income to service the debt, and growth raises the denominator of the debt-to-GDP ratio. Borrowing to finance consumption (recurrent spending, subsidies, public-sector wages) is far more dangerous — it creates an obligation with no corresponding income-generating asset, so the debt must be serviced from future taxes on an economy no larger than before. The 'golden rule' (borrow only for investment, fund current spending from current revenue) captures the distinction, though it is harder to apply than it sounds (what counts as investment is gameable). The key question to ask of any borrowing: will it generate the means to service itself, or merely the obligation?

Where the case is bounded

Two cautions bound the legitimate case for borrowing. Ricardian equivalence (Barro) is the theoretical proposition that debt-financed and tax-financed spending are equivalent because forward-looking households save in anticipation of the future taxes needed to service the debt — implying deficits don't stimulate. It mostly fails in practice (households are not that forward-looking, are credit-constrained, and don't fully internalise future taxes — especially the poor), which is why deficits do have effects, but it is a useful reminder that debt is deferred taxation, not free resources. And the deficit bias (Public Choice course): the legitimate, bounded case for borrowing (temporary, for investment, reversed in good times) collides with a political system structurally inclined to over-borrow (concentrated benefits now, diffuse costs later, on a future government). So the economics says 'borrow wisely and temporarily'; the politics tends toward 'borrow persistently' — which is why the rest of this course is about the trouble that follows.

Exercise

A government is choosing how to finance four things: (a) rebuilding after a severe flood, (b) a new deep-water port with strong projected freight revenue, (c) a permanent increase in public-sector salaries, and (d) a temporary fuel subsidy during an oil-price spike. It proposes to borrow for all four. (1) Apply tax-smoothing and the good-debt/bad-debt distinction to each. (2) For which is debt finance most and least justified, and why? (3) Explain how the intertemporal budget constraint disciplines the total. (4) Connect the salary decision to the deficit-bias warning.

Key takeaways

  • Governments legitimately borrow to smooth taxes over temporary spending spikes, finance long-lived investment, stabilise the cycle, and bridge within-year timing — debt is a tool, not inherently bad
  • Barro's tax-smoothing: hold tax rates roughly constant and use debt to absorb temporary fluctuations — so deficits should be temporary and reversed, not permanent
  • The intertemporal budget constraint: debt today must equal the present value of future primary surpluses — borrowing is deferred taxation, and a path without credible future surpluses is unsustainable
  • Good debt finances productive investment that generates the means to service itself; bad debt finances consumption (recurrent spending, subsidies) with no backing asset — the 'golden rule'
  • The legitimate, bounded case (temporary, for investment) collides with the political deficit bias toward persistent borrowing — which is the source of the trouble the course then analyses

Further reading

  1. 01

    On the Determination of the Public Debt

    Robert Barro · Journal of Political Economy 87(5) · 1979The tax-smoothing theory of public debt. The foundational economic case for deficits, and its limits.

  2. 02

    The Returns to Public Investment in Developing Countries

    IMF Fiscal Monitor / various · IMF · 2014On when borrowing for investment pays off — and the conditions (good project selection, the PIM course) for good debt.

  3. 03

    This Time Is Different: Eight Centuries of Financial Folly

    Carmen Reinhart & Kenneth Rogoff · Princeton University Press · 2009The long history of sovereign borrowing and debt crises — context for why 'good' borrowing so often turns bad. Read the whole course alongside it.

Loading progress…
LeadAfrikPublic Economics Hub