What is a technical denial in an SSA disability claim?

A denial for a nonmedical reason, issued before anyone decides whether the claimant is disabled. SSA reports them as a separate column in its outcome tables. For Title II claims the most common reason is an insufficient number of recent work credits, which is SSA's own wording in the notes to Table 60.

How many disability claims are denied without a medical decision?

Of the 2,028,572 disabled worker claims filed in 2019, 762,349 were technically denied and 4,309 were still pending at SSA's mid-2024 data cut. That is 37.7 percent of everything decided. The 2023 cohort already stood at 784,997 technical denials against 1,672,675 applications filed.

What does it take to be insured for Social Security disability?

Twenty quarters of coverage in the forty-quarter period ending with the quarter the waiting period begins, plus fully insured status in that quarter. POMS RS 00301.120. There is an alternative test for people disabled before age 31. The last day the test is met is the date last insured.

Can an SSI claimant win medically and still get nothing?

Yes, and it is common. Of the adult SSI claims filed in 2019 that reached a medical allowance, 79,077 out of 503,970 were then denied for a nonmedical reason. SSA says the most common one is excess income. The resource limit is $2,000 for an individual and $3,000 for a couple, and POMS SI 01110.003 notes that neither figure has changed since January 1, 1989.

Where do these numbers come from?

Table 60 of SSA's Annual Statistical Report on the SSDI Program, 2024, and Table 69 of the SSI Annual Statistical Report, 2024. Both come out of the agency's Disability Research File. The allowance rates I computed from the raw columns match the rates SSA prints beside them, which is the check that the columns are being read correctly.

Denied before SSA looks at the medical file

Of the disabled worker claims filed in 2019, Social Security has now decided 2,024,263 of them. It denied 762,349 without ever deciding whether the person was disabled.

Nobody weighed a consultative exam or went looking for a listing, and no vocational expert was ever called to recite job numbers down a phone line. The claim closed on a question about work history, and 37.7 percent of that year's decided claims closed the same way.

SSA calls these technical denials and gives them their own column in Table 60 of the annual SSDI report, which is where I found this. The note under the table says what is mostly driving them: an insufficient number of recent work credits.

The gate is insured status and it is not forgiving

To draw disability benefits on your own earnings record you need twenty quarters of coverage inside the forty-quarter window ending when your waiting period starts, and you need to be fully insured in that quarter as well. That is POMS RS 00301.120, and there is a separate, easier test for people who become disabled before 31.

The part that does the damage is the window. It slides. Stop working, and eventually your last covered quarter falls out the back of it, and the day that happens is your date last insured. After it passes you can get as sick as you like and it will not matter, because the question SSA asks first is not about your health.

Which is why a technical denial cannot be repaired with better medical evidence. There is nothing to argue. The earnings record says what it says, and no amount of work on the file changes the answer or produces past-due benefits to take a fee from.

In 1999 the share was one in eleven

I went back through the same column year by year, because a number that big usually turns out to be a definitional artifact.

In 1999 SSA technically denied 102,305 of 1,167,650 worker claims, or 8.8 percent. By 2004 it was 28.5. By 2009, 32.3. By 2014, 36.6. Then 37.7 for the 2019 cohort. Four times the share of applicants turned away at the door inside twenty years.

The 2023 numbers are not settled yet, since 186,886 of those claims were still pending at SSA's mid-2024 data cut, and pending claims skew toward the medical side. But 784,997 had already been technically denied out of 1,672,675 filed. Whatever the final figure is, it starts at 46.9 percent of everything filed that year and goes up from there.

My first guess was that fewer people are covered now, and that guess is wrong. The Chief Actuary's estimate of workers insured in the event of disability went from 135.8 million at the end of 1999 to 156.3 million in 2019, and 161.4 million in 2024. The covered pool grew by about 15 percent while the share of applicants failing the coverage test roughly quadrupled. So the people applying have changed, or the way they arrive at the application has, and the published tables do not say which. I could not find anything that explains it and I am not going to pretend otherwise.

SSI fails at the other end of the file

The SSI table has the same layout and a completely different shape.

Of the adult SSI claims filed in 2019 that got as far as a medical allowance, 503,970 in total, SSA then denied 79,077 for a nonmedical reason. Nearly one in six. These are people who went through the whole sequential evaluation, satisfied an adjudicator that they were disabled, and received nothing.

The equivalent column on the Title II side, same year, is 1,671 out of 643,952. About a quarter of one percent. Sixty times smaller.

The mechanism is money rather than medicine. SSA's note on the SSI table says the most common nonmedical reason for denial is excess income, and behind income sits the resource limit: $2,000 for an individual, $3,000 for a couple. POMS SI 01110.003 prints the whole history in one small table, five increases between 1974 and 1989, then a line reading "The statutory limits have not changed since 1/1/89." Thirty-seven years of prices moving and the number sitting still.

What this table cannot tell me

It lumps SSI-only claims together with concurrent ones, so some of those 79,077 are people whose Title II award is exactly what pushed them over the SSI line. Those are not losses at all, they are wins with a fee attached to the other half of the claim.

I cannot separate them out of this table and SSA does not publish a version that does. The nearest thing to a bound is at the initial level, where 628,149 of the 1,114,644 adult SSI medical decisions on 2019 claims were on concurrent applications. So the contamination could be most of that one in six or a modest slice of it, and I would be guessing either way.

The other half of the finding does not have this problem. A technical denial on a Title II claim is a technical denial, and 762,349 of them is a count, not a rate I derived.

Two gates, one of them growing

The argument in this practice area is almost entirely about the middle of the file. Which judge drew the case, what the vocational expert said, whether the listing was met. That is where the skill is and it is where the interesting fights are.

Meanwhile the outer two gates decide an enormous amount and neither one is medical. The front gate now closes on something between a third and a half of Title II applicants before an adjudicator opens the folder, and it has been widening for twenty years. The back gate takes a share of the SSI wins on a dollar figure that has not moved since January 1989.

Both are answerable from an earnings record and a conversation about money, at the start, on the first call, before anybody has read a page of medical records. That is a boring thing to be true. It is still what the agency's own outcome tables say.

The tables are Table 60 of the Annual Statistical Report on the SSDI Program, 2024 and Table 69 of the SSI Annual Statistical Report, 2024. Every row above is arithmetic on published columns, and the way to confirm you are reading the columns the same way I did is to compute the allowance rate yourself and check it against the one SSA prints in the last column. Mine matched to the tenth of a point on every year I tried.

We build the back office for disability firms, so this reaches me as a question about what gets checked before a file is opened. On the money side of the same practice, the gap between the fee cap and what representatives are actually paid is worth a look, and we pulled a year of ALJ decisions to see how much of the middle of the file comes down to the draw.