What’s an algorithm, what’s the problem?

I was at re:publica the other day and someone confidentially asked me: what exactly is an algorithm. Why is Frau Merkel herself upset and concerned? So here’s the understandable explanation.

In the old days, decisions of all sorts were made wholly by humans. If you applied for a loan, for instance, the bank would look at your proposal, but also look at who you were. Had they heard of you or your family? What is your background? And so on.

But then something new happened.

Continue reading

Why the government is right to subsidise first-time buyers

The government is not giving a grant to first-time homeowners in order to make them more affordable. Every economic and political correspondent should print this out on an A2 sheet and stick it above their bed.

The purpose of the grant (or any subsidy) is to increase the supply. In the economic jargon, this is ‘moving the supply curve to the right’. And subsidising new housing by five percent should have that effect, and reasonably fast. Continue reading

Budget gobbledygook

Another year, another budget. However, budget figures are largely impenetrable.

Tax is a great thing. We are all lucky to live in a society where we trust each other enough to contribute to a centralised pool of services that are beneficial to everyone and where taxes are collected and spent by a democratic government. Continue reading

7 impossible things before brexit

To go ahead with ‘hard Brexit’ in good conscience, the British government needs to believe all of the following:

  1. That introducing trade tariffs on trade with neighbours can make Britain into a great trading nation
  2. That the EU will let the UK sell goods and services into Europe without agreeing to follow European rules
  3. That international trading partners will slight the EU in order to hold preemptive trade negotiations with Britain
  4. That London’s banking centre can thrive without full access to the European Union
  5. That foreign manufacturers like Toyota and Nissan are too firmly entrenched to move out of Britain.
  6. That the UK isn’t dependent on EU labour to any great degree
  7. That Scotland will stay in the Union no matter what

Leader writers at the Telegraph might pretend to believe some of these things two or three days a week, but even the most hardened eurosceptic knows that at least some of these things are definitely not true. If even one of these things turns out not to be the case, it will mean a major recession at best, and the collapse of the United Kingdom at worst.

Buns, parking and the modern acute hospital

I feel for the people who run hospitals. Hospitals are mainly judged by trifling aspects of their outward presentation rather than how fundamentally good they are as hospitals. In a recent Facebook discussion I witnessed, participants got very angry about the cost of a bun in the cafe. The issue of car parking also caused consternation. Continue reading

Why water meters were a bad decision (still)

Lots of spin going on today about why water meters are a good idea and how many leaks they are finding. Apparently, Irish Water has already identified 20 houses that were using loads and loads of water. But this is only one million litres a day of water.

That is obviously great, but greater savings, and a lot less aggravation could have been had by taking much simpler and less expensive measures in relation to finding leaks, and by using the 600 million euros to replace water pipes. The 539 million euros being spent on this metering program is enough to replace thousands of kilometres of water mains (and incidentally, meter boxes could be inexpensively fitted at the same time as doing this work).

Continue reading

Irish Water has failed. What is Enda Kenny going to do now?

Irish Water has failed. That is, it has failed to get the confidence of Irish people. Here is what the government should do to solve the problem.

1. There should be a New Water Company to replace the old one. We need a new one, built on modern, sound, civically minded strategic intent and values (not like Irish Water, which was founded on rubbish like this).

2. The new water company not only needs to leave behind the failures of the Irish Water company, it also needs to leave behind the disastrous legacy of local authority controlled water, and this needs to be clearly explained. The costs, highlighted in an ESRI report, need to be cut, and the quality of service needs to be sorted out.

3. The government should stop the meter installation program immediately. The contractors who are committed to it should immediately be diverted to replacing water pipes and fixing whatever leaks they come across, whether these are inside, outside or on boundaries. This will not be straightforward, but it will make things a lot easier politically.

4. The collection of PPS numbers should be stopped and an alternative found.  Instead, the government should give people a tax rebate or social welfare supplement when they have paid their water charges. This can be done with a voucher, by direct deposit to Irish Water from DSP or Revenue, or by any number of other means. This will solve a lot of problems.

5. The government should take the management of Irish Water off Bord Gais (now rebranded as ervia). One option is to make it free-standing. Another would be to find another corporate ‘home’ for it.

6. The New Company needs new management. The management should be focused on providing good customer service. It also needs to be able to get the confidence of financiers. The current management has failed to do this, as it has failed to build cashflows. Engineering should be only the enabler. The engineering has to serve the customers, not the other way around.

7. The New Water Company should incentivise people to save water. This should be done through the voluntary installation of meters for people who think they will use less than the amount required by the flat rate. This can be made available to all users. Meters can be installed in places where the residents can see them easily and so can manage their use. Prepaid water meters should be made available where appropriate.

8. The New Company should focus its efforts on fixing pipes and getting great customer service.

9. The government needs to shut down the discussion of privatisation. In reality, there is no chance of privatising Irish Water. It is completely unsuitable for privatisation, in the same way that the gas network and the electricity network have been passed over again and again for privatisation. (And I guarantee you, the electricity network is a much more attractive candidate for privatising than the water system will ever be.) The government can do this by addressing it in the memorandum and articles of the New Water Company and in legislation.

10. The government should do all this between now and Christmas. There is really no time to waste. We need a water company we can have confidence in by the beginning of next year when the bills need to be issued.

And don’t keep on trying to do something that has been rejected. Going ‘heavy’ on people to try to get them to pay the charge is not going to work. It is just going to make everything worse politically. The government needs to accept the problems, get up, dust itself off and try again.

The political price of complexity.

When will politicians learn the political price of complexity?

Look at Irish Water.

First things first. Establishing a national utility out of a local authority patchwork of pipes and personnel is complicated to begin with. There are a lot of people to satisfy, a lot of metaphorical buried bodies, and a whole load of politics.

But layering on even more components makes the project even more complex. Look at how Irish Water has made a complicated assignment into a labyrinthine project.

Firstly, there’s the accelerated meter installation program. The plan was to have a large proportion of the two million remotely readable domestic meters installed by the beginning of this month. That didn’t happen. Installing and commissioning one of these meters requires many steps and a lot of things can and do go wrong. Dodgy pipes, bad workmanship and shared water supplies are just some of the obstacles. At the wider system level, these remotely readable meters are a new technology in Ireland and there is relatively little expertise in making them work. There are no statistics that I know of on how far Irish Water has gotten with its meter project, but it looks like they are less than half-way there.

There was really no need to install such complex meters on the first go out. A simpler in-home non-digital meter which water users could opt into having installed would have been cheaper and faster. The fancy meters could have been rolled out over a number of years, as the mains got upgraded.

Secondly, there is an ultra-complex system of charging, based on a structure of allowances and estimated assessments of water use (these assessments are critical because the meter program is so far behind schedule). The official description of the charging structure is some 31 pages long.

The simple alternative to this charging system and was described by Richard Tol in 2011. With this system, meters would be opt-in. The same allowances for children and usage could have been provided through increases in social welfare and through adjusting tax credits to put more money in people’s pockets to pay the charge.

Finally, the two factors above have converged to create a requirement for an extremely complex IT system. The requirement to collect personal information to assign allowances to individuals means that there are extremely complex issues around privacy, which Irish Water does not seem to have fully grasped.

This complexity has also resulted in a pretty odd website, a disastrous PR campaign based on some pretty dodgy conceptual work described in this fascinating PowerPoint presentation. There are ‘tribute’ sites inspired by and devoted to the topic of Irish Water and which give (un-useful) tips on what to do if Irish Water should happen to try to install one of their new fangled gadgets outside your house. The long term comedic value of the Irish Water ‘ecosystem’ should not be underestimated.

The end result of this is that Irish Water is wasting a lot of money on effort on a metering system that doesn’t really do anything to help it achieve its goal. Instead of focusing on driving down the cost of water, encouraging water saving and fixing the leaks in our rather dodgy national water network, Irish Water has gotten itself bogged down in an expensive, unpopular and ultimately unnecessary project which is going to deliver a lot of problems and very few benefits. But it isn’t Irish Water who will have to bear the cost. It will be citizens, and their elected politicians, who will pay the price.