America is experiencing the hard slog of recovering from the financial crisis. Prospects have turned more positive over the past two months. But a year ago growth was picking up too - and then it stalled, at about the same time Greece's fiscal problems infected the global economy. The question now is whether a home-grown fiscal crisis could derail this year's rebound.

Some analysts have reached dramatic conclusions, suggesting the near-certainty of hundreds of billions of dollars in government defaults within the US over the next 12 months. Such predictions will undoubtedly turn out to be substantially overblown. Yet the rejection of one extreme is not the affirmation of the other. International investors would be wise to pay close attention to fiscal trends within the US.

The severity of fiscal risk varies considerably depending on which level of government is under discussion. At the federal level the combination of ongoing weakness in the labour market and large structural budget deficits means that the right policy mix should be more stimulus now and much more deficit reduction, enacted now, to take effect in two to three years. Policymakers have acted on the first part, most prominently through the payroll tax holiday announced in December - one of the factors making the short-term outlook more promising.

They have not, however, undertaken the harder work needed to reduce projected deficits over the next decade. Most fundamentally it is difficult to see how the medium-term federal deficit can be reduced to sustainable levels without additional tax revenues from those earning less than $250,000 a year. And yet it is equally difficult to see the political system embracing that reality without being forced to do so by the bond market.

If policymakers will not act before we have a fiscal crisis at the federal level, a fiscal crisis we will ultimately have. Until then we will see a microcosm of this broader problem arise dur­ing debate about increasing the federal debt limit, later this spring. This will be contentious. We may have to experience some temporary market turbulence before it is resolved.

At the level of state governments, revenue remains more than 10 per cent below pre-recession levels. Public pensions are also significantly underfunded, leading to well-publicised concerns about debt defaults. As a recent paper from the Centre on Budget and Policy Priorities correctly argues, states will have to take immediate and painful action to reduce their operating deficits, while also gradually closing their pension gaps. Outright defaults are not likely, but these fiscal problems require concerted effort and political will, especially in larger states such as California and Illinois. This will impose some macroeconomic drag on the US economy as taxes are raised and spending is cut.

Facile analogies to indebted European countries, or the US mortgage crisis, however, are both misplaced. Although comparisons of debt across different levels of government are fraught with difficulties, US state debt levels are nowhere close to those of Greece. Debt service obligations are similarly much lower.

Unlike in the mortgage crisis, state debt has not generally been repackaged into opaque, complex securities. Furthermore, and contrary to what many pundits suggest, state governments cannot simply declare bankruptcy. Bondholders are also privileged creditors in almost all states. It is thus difficult for states to default: they would generally have to stop paying employees before they stopped making debt payments.

At the local level, however, the situation is different. Many US cities can declare bankruptcy - and given their numbers a severe crisis in at least one major city is both feasible and quite possible. As a thought experiment, take the top 30 or so cities. Assume any one has only a 2 per cent probability of a severe problem. Then the probability that at least one experiences a crisis is almost 50 per cent.

In such a city-level crisis, the state government could help - as has already occurred in Harrisburg, Pennsylvania. States would be wise to consider in advance their options in this kind of crisis scenario. But even if the relevant state government decides not to step in, and a city is forced to default, the direct macroeconomic consequences are unlikely to be substantial - unless that default triggers others to follow. In this scenario the possible contagion effect among investors in the debts of different cities is a crucial consideration.

The bottom line is that there may well be US public debt tremors this year, both during federal debate over raising the debt ceiling and with at least a limited number of crises in local and city governments. The bigger problem, though, lies beyond 2011, as the unsustainability of the federal government's fiscal trajectory becomes increasingly clear. I hope it does not ultimately require a crisis to restore fiscal sustainability at the federal level, but I fear it will.