21 November 2013

It seems beyond belief that the nation that brought the world Amazon, Google and Facebook is seeing the president’s flagship reform founder on the rock of a defective website. The current wave of introspection and recrimination in the US about the problems of Obamacare will resonate with UK audiences and remind us why implementation matters.

President Obama has staked his political reputation on the Affordable Care Act – designed to bring health care to millions of uninsured Americans. He pushed it through Congress on the most acrimonious of split votes, with not a single Republican voting in its favour. Had Romney won last year, Obamacare would have been consigned to the legislative dustbin. Its durability – and the president’s legacy– depends on getting off to a good start and becoming an entrenched part of US entitlements by the time power next changes.

Yet the public debut of Obamacare has been a disaster, making our government’s local difficulties over universal credit look trivial in comparison.

First, people who want to register are finding it very, very hard. Health and Human Services Secretary Kathleen Sibelius found herself testifying to Congress that the website was slow but functioning – alongside a split screen saying the site was down. The IT systems required to administer Obamacare take in six central government departments, state governments and vast swathes of the healthcare sector, from practitioners to private hospitals. The complex co-ordination problems inherent in such a system are now manifesting themselves as a major customer service failure.

Second, not enough of the right sorts of people are trying to register – undermining the fundamental economics of Obamacare which depend on it being more than a refuge for the hard to insure. In this the president has had to contend with a Supreme Court decision which allowed states to opt out of one of the supporting elements of Obamacare – the expansion of Medicaid (funded healthcare for the poor).

Third, the president has been called out for his reassurance that millions of existing healthcare policies which do not meet a minimum standard set by the Act, will not be immediately cancelled if people want to keep them. Former President Bill Clinton entered the fray, demanding the president honour his pledge – and emergency legislation is being pushed through to create a year’s breathing space. But that looks like postponement not solution. Inflict the pain now, or wait 11 months? Neither is particularly palatable for the president or the Democrats facing election campaigns in 2014.

The consequence is that the president’s trust ratings have plummeted – which puts at risk a key source of presidential authority. While the reputational damage to the commander-in-chief himself is substantial, it has also bled onto the party. Recent polls show the gains that the Democrats made by successfully pinning the government shutdown on stubborn Republicans being eroded by Obamacare’s failures. Not content with having to battle a partisan Congress, Obama must now fight to rebuild the public confidence in his initiative.

There is no shortage of analysis of what has gone wrong. An article by Levin and Shah in Foreign Affairs argues that the core problem is inflexible implementation in secret, with no accountability and an absence of expertise:

"The website was an engineering problem attached to a political program, Obamacare. Rather than solving it like engineers, though, officials tried to solve it like politicians. They hid the strategy (and the software) in the naive belief they were capable of delivering a complex system in a technical vacuum. This approach is toxic when it comes to large-scale implementation where one wants more transparency. The political decision which wreaked the most harm was to insist that a massive social program could be rolled out across the country on a single day without experienced management, detailed reporting, comprehensive testing and a workable plan for implementing the whole system."

They go on to argue that the ‘choose once’ approach to public procurement cannot cope "when the service is something unprecedented, save healthcare.gov for which the final specifications were, by the very nature, of the program, only vaguely understood at the outset". That sounds very much like a call for a more agile approach to solving a new problem and the sort of transparency we lauded as a hallmark of the successful delivery of London 2012.

But did no one tell the president? One of the NAO’s criticisms of DWP’s performance on universal credit was a reluctance to transmit bad news upward. In the US system, with the top ranks of government staffed with hand-picked political appointees, surely someone had the authority to tell the president he was risking disaster. Not according to CNN’s Gloria Borger who claims the problems were caused by silo working with "no roll-out czar" who could pull the whole thing together and a president who likes "everyone to stay in their lane", compounded by a culture of non-disclosure, described as "if you hear something outside your purview that is bad, you may not want to report it; everyone thinks that if it isn’t my lane, and I talk about it I’m being a tittletattle". Under this analysis, the president’s desire to avoid internal dramas meant he wasn’t warned he was risking a full-scale crisis, and no one saw it as their job to tell him he was. This underlines the need for internal openness and clear accountability.

The president is now issuing mea culpas by the day. Iain Duncan-Smith might note that he is not laying the blame on his officials – in public at least. The administration will – eventually – solve its website problem. The political and economic problems – resulting from introducing major reforms in a highly polarised system, the role of both Congress and the Supreme Court in rewriting the initial intents of legislation – are potentially much more damaging in the long-run. But the case of Obamacare is a timely reminder of how much getting implementation right matters – which is why the Institute for Government is now undertaking a major project to look at this. But it is also a useful reminder that messing up implementation is not a uniquely British phenomenon.


It is cold comfort to know that other administrations can waste money on computerised systems just as much as the UK. I.e. no comfort at all.

The problem of Universal Credit remains and needs to be solved.

Everyone accepts that the politicians, the officials, the consultants and the contractors failed with Universal Credit. Whatever they did, we don't want to do it again.

So we need to do something else to make work pay and to stop the expensive churning of money backwards and forwards between the benefits systems and the tax systems.

There is a vociferous claque in favour of so-called "agile" systems engineering. That's something different. Perhaps, like a magic wand, agile will work where other methodologies have failed.

The National Audit Office very sensible printed the entire manifesto of the Agile Alliance in their 'Universal Credit: Early Progress' report. Not just the manifesto, but the 12 principles behind it. The whole lot fits on to one page. There's much to it. It dates back to 2001. If it was a magic wand, agile would have swept all before it in the past 12 years. It hasn't and it isn't.

DWP's internal report, presumably Mr Howard Shiplee's report, was leaked to the Guardian. According to the Guardian article, DWP believe they have to choose between (a) sticking with the existing contractors and making Universal Credit work and (b) starting a new agile Universal Credit project. They may opt for (a) and (b).

The risks associated with (a) and (b) are so great that the clearly correct choice is neither (a) nor (b).

That much is clear.

Nothing else is.



Agile v. digital-by-default, http://www.dmossesq.com/2013/11/agile-v-digital-by-default.html

Universal Credit and GDS – think twice, http://www.dmossesq.com/2013/11/universal-credit-and-gds-think-twice.html

This approach was taken because the president and congressional leaders had concluded that more progressive plans, such as the (single-payer) Medicare for All act , could not obtain filibuster-proof support in the Senate. By deliberately drawing on bipartisan ideas — the same basic outline was supported by former Senate majority leaders Howard Baker , Bob Dole , Tom Daschle and George J. Mitchell —the bill's drafters hoped to increase the chances of garnering the necessary votes for passage.

Indeed there was a strategy being developed in the late 1990s under the Information Age Government Champions, a cross-departmental group of Civil Servants (I had a contract to edit the Framework for Smart Cards in Government). But the impetus to deliver was lost at the start of 2005 when Blair pulled his project to upskill Whitehall, as confirmed in the recent BBC File on 4 programmes 'In Defence of Bureaucracy'. Again I had a contract, this time in 2004/5 associated with smart ticketing for public transport, pulling together input from various sources including CESG, but it was cancelled in early 2005.

Add new comment

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.