How $55.7 Million Doesn’t Equal $634 Million
There is no denying that the rollout of the Obamacare exchange website has thus far been a complete and utter disaster. I tend to agree with the assessment that if these problems aren’t fixed – and soon – it will start to ensure that the “young invincibles” that Obamacare needs to have any chance at addressing the adverse selection problem won’t sign up.
However, there’s a story making the rounds claiming that not only is the website a complete failure thus far, but it has suffered from cost overruns on the order of 1000 percent, with an initial award of $57 million (and additional option years bringing the potential value to a little less than $100 million) but payments to date of over $634 million. This story has taken in people who should know better.
It’s also very wrong, based on a poor understanding of government contracts, and in particular a poor understanding of how data on government contracts is reported. That’s not to say I’m an expert myself- the world of government contracts law is a dark and mysterious place with its own lingo that seems to take people years to learn. Hell, it even has its own special court. I do, however, know a minimal amount about the field – enough to know that one really needs to do one’s homework before jumping to any conclusions about a given contract.
Here, the author of the original story – who appears to be an Obamacare supporter – appears to have based his conclusion on the data from this page of the government’s spending website, which filters all awards under the relevant contract number. And indeed it shows that $634 million has been awarded (not necessarily paid, since all amounts allocated by a contract are allocated to the year of the original award, not the year of payment) under the contract number that includes the exchange website. The amount of the original award for construction of the exchange website, $55 million for the first two years, with three one-year options that could bring the total value to $93 million. That contract was awarded in December 2011 – keep that date in mind.
But if you look a little bit more closely, you’ll start to see that the “contract number that includes the exchange website” is not the same thing as the “contract for the exchange website.” For instance, take a look at the “List of Transactions” for this contract number. You should quickly notice that the award dates on most of the line items are from before December 2011 – in other words, those line items, which are included in the aforementioned $634 million number, were incurred before the exchange website contract was even awarded. Except for the second line item (which happens to be for $55.7 million, the exact amount of the original award), you’ll notice that the other line items list a variety of award dates in 2012 and 2013, suggesting they’re different awards from the exchange website award (hint: they are).
Notice also the charts at the bottom of the page from which the $634 million figure was originally pulled. Those charts list three separate prime award contractors; although each is an affiliate of CGI Federal, it would be fairly odd to have three separate prime awards of the same contract; again, the implication is that the $634 million covers far more than just the exchange website contract. Next, look at the graph at the bottom of the page, which shows awards dating back to 2008, with $152 million awarded in 2011. It seems rather unlikely that between September 30, 2011 and December 31, 2011, CGI Federal was paid $152 million on a $55.7 million contract, no? Besides, as I mentioned above, contract amounts are attributed to the year in which they were awarded, not the year in which they are paid, so if this company had really overrun the award amount as much as the original story claims, the 2011 figure should be $634 million, not $152 million.
In other words, that $634 million is for multiple awards, not just the exchange website. But, you ask, how can that be if those awards are all connected by a single contract number? Well, let’s take a look at the filter used to find that contract number, which has the mysterious letters “IDVPIID.” Remember how I said government contracts have their own weird language? Well, this is a good example.
I had to do some research (again, I’m no government contracts expert), but a little looking around quickly revealed what those letters mean: “Indefinite delivery vehicle procurement instrument identifier.” The important part of that phrase for our purposes is the “indefinite delivery vehicle” part – that phrase is legally defined as an “agreement or contract against which individual orders or purchases may be placed.” In other words, the company that built the website is what might be called an “approved vendor” in other contexts, and is subject to an additional long-term “master” agreement – and has been for quite some time. That doesn’t mean the project wasn’t competitively bid – according to the contract data, four bids were received for the project – just that the project was awarded subject to the terms of the long-term agreement, and probably only similarly approved vendors were eligible to bid (though I could well be wrong about that last part).
So if the $634 million claim is inaccurate, how much has actually been spent on this website to date? The answer is that we don’t know, at least not exactly. But we do know that the data for the website award still lists the “current contract value” at the original $55.7 million. That number may increase next year and the following two years if the government exercises its renewal options, but the key point is that the obligation amount does not appear to have changed even though the initial period of the contract is almost at its end. If more than that amount has actually been spent, there’s not currently any evidence to support such a conclusion. But we can say with a good degree of certainty that the amount spent on the project is nowhere in the ballpark of $634 million, which is no more and no less than the total amount awarded to this particular company for over 110 projects since at least 2010 and possibly earlier.
We should absolutely be vigilant about government waste, and I’d hope the government would get better value for $55.7 million than it has so far, particularly as the stakes of getting this wrong are far higher than a $55.7 million contract. But $55.7 million isn’t $634 million.
UPDATE: It is worth mentioning that the above-referenced amounts are not the sole contracts for creation of the exchanges, as this GAO report indicates (via), and the total costs of the exchange creation contracts do reach into the hundreds of millions of dollars. But that is different from the claim being addressed in this post, which is that there were cost overruns in excess of 1000% on this one particular contract for the creation of the healthcare.gov website by CGI Federal.
UPDATE 2: it appears the original article at the center of all of this has been replaced by a new article that speculates about the total costs of the exchanges, which as I say above is a very different issue than the cost of this particular contract. The original claim appears to have been scrubbed and memory holed, evidenced only by a note indicating that the previous article had a “miscalculation,” but making no reference to the fact that the previous article was on a different topic altogether, nor identifying well the nature of the “miscalculation.”. I have various other problems with the replacement article, but ultimately, it’s pretty evident that much of what is in the replacement article is speculation. The proper way for the author to handle this would have been to simply place a retraction at the top of the original article along with a note that the author is investigating the related story of how much the exchange website has cost or is likely to cost.