Appraisal Optimism

There are plenty of examples of poor planning of IT and infrastructure projects – and such failures are equally common outside government - think BAA/BA's Terminal Five and TSB Bank.  (See also Note 1.)  The big difference is that public sector failures are very public.  Private sector ones are usually pretty well hidden unless they impact passengers or bank customers.

One problem is that it is all too easy for both those involved in planning large projects to be tempted into displaying what the World Bank tactfully calls appraisal optimism – the presentation of best case scenarios including under-estimated costs and timescales, and over-estimated benefits, in order to get projects started. 

Examples include

And appraisal optimism is not just a British disease.  The Sydney Opera House was completed in 15 years, rather than five, and cost A$102m rather than A$7m.

These examples all perhaps support the seductive proposition that nothing would ever get built without a little massaging of the facts and figures, but such behaviour all too often leads to later criticism of the officials responsible for the project, rather than the politicians who pressed for it to proceed.

Presidents and Prime Ministers are of course under great pressure to announce dramatic goals and projects.  Some of these work out OK, such as some of those listed above, and Kennedy’s 1961 belief “that this nation should commit itself to achieving the goal, before this decade is out, of landing a man on the moon and returning him safely to the Earth”.  Neil Armstrong stepped onto the Sea of Tranquility in 1969. 

Some don’t turn out quite so well.  UK examples include numerous defence and IT projects, including Universal Benefit. Politicians are wont to complain that civil servants are too risk averse and oppose all change, including major projects. But the National Audit Office does not believe that officials offer deliberate resistance to Ministers’ demands. On the contrary,, there is ‘unacknowledged reality’ as a result of which officials often ignore inconvenient facts. The NAO’s view is not that civil servants are risk averse. It is that civil servants too often do not understand the risks attached to what they want to do.

In the case of Universal Credit, Robert Devereux, the responsible Permanent Secretary, was quoted as follows by Civil Service World’s Susannah Brecknell:

“It’s not a state secret that the views of my ministers and the views of other ministers about this project were different.” In this environment, he says, people’s desire for candour can be affected by “high-level politics”, suggesting that people kept challenging discussions in closed spaces to keep the project going. Devereux now accepts that the two fundamental elements of the Universal Credit roll-out were undeliverable with the time and resources available and should not have been attempted at the same time.

In reality, how easy is it for an official to say: “Minister, you have all these priorities – which one first?”  Devereux suggests that the more realistic message might be to warn ministers that an ambitious programme of projects is bound to result in complications, and he says it isn’t only perm secs who can make this point – non-execs already do, and select committees could make it more.

“All of the non-execs we’ve ever had, having been here for five minutes turn round to the secretary of state and say, ‘we would never do as much change as you’re doing’ and typically the answer they get is, ‘yeah but we’re government’,” he says.

More specifically, the NAO has listed the following problems which all too often lead major projects to under-deliver on benefits and to overshoot on time and cost.

Ian Walmsley makes similar points when he draws attention to the optimism squared rule in his Confessions of a Commissioning [Railway] Engineer. Here are some extracts, slightly rewritten for the purposes of this web page:

My own experience of commissioning is pretty limited, but that didn’t stop me laying out numerous optimistic Gantt charts over the years, saying how long things would take. In my defence I used to make sure suppliers agreed, but of course I was speaking to their sales team who as a rule don’t know any better. It’s their job, so they will often half the timescale that their production team have promised, if that is necessary to win the deal.

The Minister supported by his/her senior officials will separately be looking for better performance and/or lower cost, as he or she needs that to make their case to the professional sceptics in the Treasury. Without it, the department might as well keep what the current systems - so ‘I want to believe’ takes over. Most of us experienced officials were more likely to believe in Jedi Knights than sales people, but expectations in today’s officialdom rise in inverse proportion to experience. Consequently, the customer will be looking to double the performance he has now.

So doubled promises reinforce doubled expectations: the optimism squared law.

Further Reading

Here is some helpful advice:


1.   Other UK examples include:

2.  And overseas:


Spotted something wrong?
Please do drop me an email if you spot anything that is out-of-date, or any other errors, typos or faulty links.