Congress may have ordered agencies that regulate financial institutions to stop relying on credit ratings by firms such as Moody’s and Standard & Poor’s, but regulators are still struggling to come up with an alternative.

One of the painful lessons of the mortgage meltdown was that securities stamped with the highest ratings turned out to be poor investments.

Yet those private ratings had been enshrined in the nation’s laws and regulations as official benchmarks of quality, helping to determine, for example, what assets certain financial institutions could buy and how much money they needed to hold in reserve in case the securities imploded.

In its legislative response to the financial crisis last year, Congress told regulators such as the Securities and Exchange Commission to go through their rules and strike any official reference to credit ratings. The SEC opened another chapter in that review Wednesday, even as some commissioners said the task could be difficult.

In solving one problem, Commissioner Luis A. Aguilar said, the agency could create others.

The five commissioners unanimously proposed a set of rules Wednesday addressing how to protect clients of brokerage firms without leaning on the familiar crutch of credit ratings.

More specifically, the rules govern how much of a financial cushion — otherwise known as capital — brokerage firms must maintain to cover obligations to investors if the firms go under.

Currently, the grades that rating agencies assign to securities the brokerages hold are used to measure the value of those securities and the size of the cushions.

In place of that seemingly objective but largely discredited yardstick, the SEC proposed allowing brokerages to use a mix of factors, and it proposed letting them tailor their own combination of measures, potentially opening the door to standards that vary from firm to firm.

In assessing the quality of a bond, for example, brokerages could consider how its interest rate compares with the yield on U.S. Treasury securities, which are considered the safest investments. A wider gap, or “spread,” between the two would signal that the bond in question carries greater risk.

Other proposed factors include “internal or external credit risk assessments,” the default rate for the securities, and whether those securities carry any form of insurance backstopping their creditworthiness.

Internal or external assessments could be subjective, and default rates might have limited predictive value. During the housing bubble, many securities backed by mortgages had low default rates — until they had high default rates.

“The dilemma that confronts the commission is what to do when no appropriate substitute exists,” Aguilar said.

“Clearly, the proposed subjective standard will be much more difficult to police than the current objective standard that references credit ratings,” he added.

Brokerage firms would have an incentive to overestimate the quality of their investments so they could operate with less capital and lower costs, he said.

Commissioner Kathleen L. Casey expressed a different view. She said ratings “should not serve as a substitute for appropriate creditworthiness assessment.”

SEC Chairman Mary L. Schapiro said the agency faced a challenging assignment.

“Although these rules are rather technical,” she said, “we must not lose sight of the fact that they form a critical part of the commission’s customer protection regime.”

The proposal the SEC advanced Wednesday is now open for public comment.

In another piece of the Dodd-Frank regulatory overhaul, the agency earlier this year issued a proposal to strip credit ratings from the regulation of money-market funds.