Regulatory creep continues to transform the relatively straightforward Community Reinvestment Act into a much broader law that aims to cure many social problems. Much like military mission creep, the program started with a specific objective but gradually became bigger than anyone ever anticipated.

Whenever I am asked about reform or a related important CRA question, I recall my many discussions and meetings with former Sen. Bill Proxmire. He is the father of the CRA who heroically pushed this law through Congress in 1977, despite considerable headwinds from other politicians, the banking industry, and even the Federal Reserve.

At the risk of sounding immodest, it is relevant to note that he recognized I had a good understanding of his law's original mission when he endorsed Community Reinvestment Performance, my first book on the CRA. Proxmire wrote in 1993:

It is far and away the best analysis of government regulation that I have seen in any field. He spotlights the regulatory problems that continue in CRA and points out precisely how they are being overcome. CRA will benefit enormously from this superlative examination and report.

The CRA requires banks and thrifts to help meet the credit needs of their entire community, including low- and moderate-income, or LMI, neighborhoods. This was originally interpreted by the regulators as lending in these geographical areas but was properly expanded to include lending to LMI people and small businesses and farms.

Proxmire recommended that then-Comptroller of the Currency Gene Ludwig consult with me on the 1995 CRA reforms, and I had the opportunity to provide direct input to him and his staff. The resultant loan, investment, service, and community development tests and exam procedures were designed to focus on meeting community credit needs.

The 1995 reforms also contained more general language encouraging community revitalization and stabilization activities as well as community and economic development. I initially wondered if the wording in this reform was too broad, but took comfort in its LMI focus.

As a CRA purist, I also questioned the practice of enforcing fair lending laws by downgrading CRA ratings, since the law is intended to be income-based rather than race-based. I later realized that credit needs in LMI neighborhoods could not be met by banks violating fair lending laws.

However, the use of the CRA as a fair lending enforcement tool was the beginning of the law's regulatory creep. Community groups and some government officials gradually discovered how to use the broad reform language to attempt to address other social problems.

This effort began with environmental issues such as brownfield redevelopment projects, which were considered acceptable community revitalization activities for financial institutions. The reform language was broad enough to allow consideration of other similarly socially desirable activities with an indirect LMI benefit to communities, rather than focusing on direct credit extension.

It was only a matter of time before community groups and government officials began using CRA rationale to address other social issues involving community facilities like hospitals and elementary schools and community services such as alcohol and drug recovery centers and youth programs. Similar arguments were used to grant CRA credit for certain job training programs, building infrastructure such as sewer lines, and demolishing blighted structures. There is even a current push to include health and nourishment activities under a community development angle.

A CRA purist might say, "Leave CRA alone, and go find another law!" But a CRA practitioner in the spirit of Proxmire should ask, "How will these other social activities help financial institutions meet LMI credit needs?" And it must also be asked whether these indirect social activities are distracting banks from their primary mission of direct LMI lending to people and small businesses.

This latter consideration is particularly important given the fragility of small businesses in our current economy. Americans are starting fewer businesses and new businesses are failing more quickly, according to a Sept. 11 hearing by the House Committee on Small Business. The remaining new firms are creating fewer jobs. For the first time in 30 years, more businesses now close than open, according to the U.S. Census Bureau.

The most recent CRA small business lending data from the Fed indicates that small business lending by reporting banks was basically flat between 2009 and 2013, as was the percent of loans (36%) to firms with annual revenues of $1 million or less.

Mortgage lending to LMI Americans has also fallen significantly in the aftermath of the most recent financial crisis. The proportion of home purchase loans made to such borrowers has decreased every year between 2009 and 2013, falling from 37% to 28%, according to the Fed's analysis of recently released Home Mortgage Disclosure Act data. The same analysis shows the portion of home purchase loans in LMI neighborhoods has remained essentially flat at 13% over this period.

Considering record-low mortgage rates over this period and the fact that low- and moderate-income is generally defined as the lowest 40% of the income scale, these numbers suggest substantial room for improvement.

Perhaps the situation can be remedied if we refocus our CRA efforts on the direct lending purpose of that 1970s law, rather than using it indirectly to try to cure our country's social ills. Proxmire cared deeply about our country's LMI communities, and his legacy CRA law will be most effective if we remember its original purpose.

Kenneth H. Thomas, an independent bank consultant and economist, was a lecturer in finance at the University of Pennsylvania's Wharton School for more than 40 years. He is the author of The CRA Handbook.