The government’s mortgage modification program has fallen short on any number of counts. Here’s yet another: It’s failed to boost the number of modifications relative to the need.

Every month, we post a graphic rundown of the new numbers from the administration’s program. Here are the latest results, broken down by each bank or mortgage servicer participating in the program.

For months, the numbers have been showing the same trend. More and more homeowners have been disqualified from the program (despite the fact that most continued to stay current on their payments). Meanwhile, fewer homeowners are entering the program.

The government’s effort is unlikely to come anywhere close to meeting the administration’s stated goals of offering help to three to four million homeowners. It also seems unlikely to spend anywhere near the full $50 billion originally set aside for the program. The number of homeowners in ongoing mortgage modifications through the program could start shrinking soon.

With the program’s reach declining, the administration has pointed to what else banks are doing to curb the problem. Banks and other companies servicing mortgages are providing far more modifications on their own: In September the ratio was more than four to one.

Earlier this month, we looked at the quality of those modifications. Although they’re better than they used to be (most used to actually raise the homeowner’s payments), the banks’ in-house modifications are on average half as generous as those the government sponsored. Homeowners in those modifications are also twice as likely to default as those with government mods.

Here’s a look at the overall number of modifications: In September of this year, servicers completed about 147,000—only about 28,000 through the government’s program and the rest outside of it, according to HOPE Now. Meanwhile, about 5.2 million mortgages were in default (60 days or more behind), according to LPS Applied Analytics. Put those two numbers together, and you get a sense for the number of modifications being done relative to the need: In September, servicers provided modifications for about 2.9 percent of the total delinquent loans.

As this chart shows, that’s about how many the industry was doing before the government’s program launched in March, 2009:

In February of 2009, the industry did about 127,000 modifications, while there were about 4.4 million delinquencies. That comes out to about 2.9 percent.

You can see a big dip in the graph that coincides with the program’s launch. That’s because servicers agreed to evaluate all homeowners for the government’s program before considering other modifications. Servicers couldn’t keep up, and a huge bottleneck resulted. They didn’t start approving significant numbers of homeowners for government-sponsored modifications until November of last year. It took servicers until this March, a year after the program’s launch, to return to their previous level. We’ve reported many times on the extended delays homeowners have suffered.

The bottom line is that for homeowners, modifications are just as rare as they were before the program launched. The absolute number of modifications is higher now than it was then, but so are the number of defaulted loans.

That doesn’t mean the rest of these mortgages have been sold through foreclosure. For the most part, loans have gone months upon months with no resolution. In January of 2009, the average mortgage in the foreclosure process had been delinquent for 319 days, according to LPS Applied Analytics. In September of this year, that number was 484. There are over 2 million loans in the foreclosure process, but there were only about 120,000 foreclosure sales in September, according to HOPE Now.

There’s likely to be a spike in foreclosures down the line unless there’s a major change. Speaking in July, Jeff Carbiener, the president of Lender Processing Services, which provides the systems many servicers use to process payments and foreclosures, told analysts on an earnings conference call that a “significant flow [of foreclosures] has to come through the pipe at some point in time.” He said there had been “some success” with modifications, but they hadn’t “taken a dent” out of the number of loans in the foreclosure process.

It remains to be seen how much of an impact the recent robo-signing scandal will have on that impending wave of foreclosures. Recently, analysts from Moody’s concluded that the problems will “cause delays rather than inability of servicers to foreclose.” The state attorneys general, meanwhile, hope to use their investigation to increase the number of modifications.