Large U.S. banking companies will post $44 billion in writedowns and credit-loss provisions in the fourth quarter, forcing them to use capital injections from the federal government to beef up their balance sheets instead of lending the money out to consumers, bank analyst Meredith Whitney predicted Wednesday.

In a widely circulated research note, Ms. Whitney, of Oppenheimer & Co., wrote that the Treasury Department's Troubled Asset Relief Program, which has been used to pump hundreds of billion of dollars into banks over the past month in hopes of freeing the flow of credit, will not work as intended.

Ms. Whitney, who did not respond to an interview request, wrote that most large banking companies, hampered by continued mortgage losses and now credit card and other loan losses, will be forced to use Tarp capital to boost reserves for 2009.

She estimated that five of the largest banking companies she covers — Citigroup Inc., Bank of America Corp., JPMorgan Chase & Co., Goldman Sachs Group Inc., and Morgan Stanley — could use up to $18.8 billion, or about 22% of the money they collectively received under the initial Tarp payouts last month, to fortify reserves in the fourth quarter.

"Overall, we do not believe these capital raises will spur meaningful growth for the industry. We remain cautious on the financial institutions as they continue to face asset price declines and a prolonged weak economic environment," Ms. Whitney wrote.

Also Wednesday, Ms. Whitney said Citi would post a 2008 loss of $3.02 a share, instead of the $2.87 she estimated previously.

At $45 billion, including a $20 billion emergency injection last week, Citi has raised more capital than any other company under Tarp.

The companies that Ms. Whitney cited either did not immediately return calls Wednesday or said they do not comment on analysts' reports.

Subscribe Now

Access to authoritative analysis and perspective and our data-driven report series.

14-Day Free Trial

No credit card required. Complete access to articles, breaking news and industry data.