Financial policymakers focused on rewriting the rulebook for financial markets have spent months debating issues related to consumer protections, systemic risk regulation and regulatory co-ordination. But none of these is as important as what to do when a large financial institution is failing.

Right now, we are no better off than a year ago, when several major institutions teetered on the edge of total failure, and nearly all of them were granted massive infusions of government and central bank funds. The bailouts were done on an ad hoc basis, as policymakers sought to prevent the catastrophic failure of a few institutions to spare theglobal economy.

But while better regulation would have helped avert some of the crisis, there is no guarantee that regulators can avoid a similar crisis in the future. We must address the lack of an orderly process for the demise of large financial institutions. Before last year, we were all aware of "moral hazard" and what it meant - we just did not know it referred to far greater swaths of the financial markets.

So in the unique series of events that started with the fall of Bear Stearns, there was no way to deliver an orderly lesson on the danger of moral hazard. As a result, investors, lenders and management could believe, in almost every case, their institutions were "too big to fail".

To avoid another crisis we must address this flaw in our regulatory system. We need a process that is known in advance, includes the efficient and orderly write-down of assets and restructuring of debt, and draws on public funds only as a last resort. We need a resolution authority created explicitly to impose discipline on those with the most power to influence the level of risk-taking: the holders of both equity and debt of institutions that may be "too big to fail."

What we must also recognise is that in our desire to remove "too big to fail" from the lexicon, we should avoid creating a greater degree of panic among investors when large institutions hit bumps in the road. While some investors are insufficiently curious about the risks large institutions take, others are happy to race for the exits at the first sign of financial strain. A resolution authority must be designed to strike a balance between managing large firm failures so as to reinforce market discipline and minimising market panics.

At its core, a resolution authority would have vast powers, in some cases, greater than the bankruptcy courts. It would not act in a mechanical way, following a playbook written in advance. Instead it would address the problems of a single institution in the context of our complex markets, where other institutions are likely to be affected by a single failure.

As defined by the Cross-Border Resolution Group, such a process would have to account for a number of factors, including that nations have different approaches to institutional failure and thosemust be co-ordinated due to cross-border financial markets.

In today's markets, we need a resolution authority that draws on a few core principles. First, it should ensure investors, long-term debt holders, lenders and management suffer pain commensurate with the risks they allowed an institution to take.

Second, following the model of the insurance required for depository institutions, create a self-insurance programme that provides a buffer and prevents panic on the part of short-term creditors and the public.

Third, do not lose sight of the overall economy - sometimes shutting down a failing institution would discourage future moral hazard, but at such great cost it is best to save it.

Fourth, work closely with the systemic risk regulator and other supervisors to head off compound failures.

Fifth, when failure must occur, do it swiftly. Retain the right to transfer contractual obligations to healthy firms, sell assets, permit intra-institutional lending, cancel debt or equity, grant special protection to certain creditors and seek penalties from management for taking excessive risk even after multiple warnings.

Think what these principles, and the actions that would follow, would mean to the daily decisions of management, customers, creditors and investors. They would become more careful - having foreknowledge of their potential rights and responsibilities should the resolution authority be activated. By having a personal stake in potential failure, they would price debt and equity differently. That would lead to a greater diversification of assets and better risk management.

These actions may do the work of many regulators. Those charged with providing independent judgments on the quality of an institution's assets - debt ratings agencies, accounting firms, analysts and regulators - would be relied upon and held accountable even more for their objective views.

Regulators cannot catch every problem, especially in the fast-moving, innovative financial markets. Sometimes, a solution that requires the demise of an institution is not practical as it would threaten so much else. The goal should not be the rough justice of market discipline in all cases, but the discipline of prudent behaviour that leads to market stability.

The writer is adviser at the Carlyle Group and Goldman Sachs and former chairman of the Securities and Exchange Commission