As the German election campaign draws to a close, financial experts are again starting to look at Southern Europe. Will the eurozone relapse into crisis once the new German chancellor is in place? While this is clearly a question of immense importance for Europe’s future, the arguments in the debate are surprisingly unsophisticated. If Europe’s leaders concentrate on nothing but debt levels, they may be tempted to take the wrong decisions.

Economists are hotly debating the question of how much debt a country’s economy can sustain before it slows down. Most recently, two economics professors, Kenneth Rogoff and Carmen Reinhart, suggested 90 percent of GDP as the cut-off point. Yet what is missing in the debate is an explanation of why any debt ratio would cause slower growth. And is 90 percent really a reasonable threshold?

It isn’t. Excessive debt, it turns out, can indeed hamper a country’s growth prospects, but it does so because economic agents respond to what they see as a rising probability of default. When a country’s balance sheet starts to look too risky, it begins to generate what experts call financial distress costs—just as businesses do when they are suspected of courting bankruptcy.

Major stakeholders, in trying to protect themselves from risk, change their behavior in ways that automatically reduce growth and increase financial fragility further. In effect, they create a feedback loop in which their behavior worsens the risks and therefore further increases the incentive they have to protect themselves.

For example, creditors raise their lending rates and shorten maturities. Businesses, worried about future taxes or even expropriation as the government struggles to service debt, protect themselves by disinvesting. Workers organize to protect their jobs and behave in more militant ways. Savers, concerned about inflationary policies that will erode the value of savings, withdraw deposits from banks. Policymakers shorten their time horizons as they try to manage the next crisis.

These stakeholders all react to the rising probability of default by taking steps to protect themselves from default. It is this highly procyclical behavior that causes the economy to slow. Even in theory, however, there is no such thing as a universal debt threshold. For one thing, if a country borrows to invest in wealth-generating projects, its debt capacity rises faster than its debt-servicing costs, in which case higher debt levels are not necessarily bad.

Also, there are at least three other factors besides the level of debt that matter. First is the economy’s underlying volatility. Countries with very volatile economies are more likely to find themselves in a situation where economic growth is no longer sufficient to allow debt-servicing. At any given level of debt, they are more likely to default, so they begin to incur financial distress costs much earlier.

The second factor is how highly “inverted” the country’s balance sheet is. This measures how fast the costs of servicing the debt rise when the economy slows. This can put further downward pressure on the economy in another self-reinforcing feedback loop. Common forms of inverted financing include foreign currency and short-term debt.

The third factor typically consists of contingent liabilities in the banking system that emerge after an adverse shock to the economy. Spain is one of the clearest recent examples of this process. Before the euro crisis, Spanish government debt was lower than that of many other European countries, but its banking system was overexposed to a real-estate bubble.

The key point is that an evaluation of financial fragility must be forward-looking, not backward-looking. If analysts are to understand how debt can hurt economic growth, it is not enough to look at whether a country can easily service debt under current conditions. Any assessment must also consider the probability of the debt-servicing costs surging because of an adverse shock.

In other words, in considering the creditworthiness of Spanish banks, economists must consider how likely it is that an unexpected external shock in the near future will cause more deterioration in bank loans. This could reduce the creditworthiness of the government, exacerbating the shock.

Financial crises usually come as a surprise because observers ignore the strong feedback loops embedded in a country’s economic institutions. In assessing the balance sheet of countries like Spain, Italy, Portugal, and others, it is not enough to ask whether the current debt-to-GDP ratio is too high. Analysts must consider how volatile the economy is, evaluate how inverted the balance sheet is, and ask what contingent liabilities are likely to emerge in the case of plausible adverse shocks.

The problems that Europe faces are not just due to high debt levels. They are above all the highly procyclical connection between the health of the economy and the evolution of debt-servicing costs. Because of this, a few weeks of good news will likely lead to a too-rapid return of optimism as all the raw debt numbers suddenly seem to improve. But the same mechanism that causes a rapid improvement in sentiment can lead to an equally rapid deterioration. Until Europe’s balance sheets are stabilized, sentiment will continue to be volatile.