Posted by: Martin | July 6, 2011

Taxi and scurvy data

A small taxi experiment

One thing with the taxi business in Africa, unlike say electrical engineering,  is that everybody seems to have an opinion about it – and often contradictory opinions.  Here are examples of two comments I have received:

I would not recommend at all to start TAXI business […] Normally all people I know who ever sent a car to Africa have never seen a cent out of the business. Its very difficult to have a business in Africa and be successful.

Taxi’s are a profitable because everything can or could be be arranged: car papers, insurance, drivers license, repairs, etc.  A friend of mine […] is planning to convert it into a woro-woro. His only problem is the reliable driver. If you have one, he says, owning a taxi is most definitely a profitable business.

My take here, inspired by author Peter Sims, is that the world is a complex and changing place, and you can’t always figure out and plan everything in advance, so a good approach is to make a small, not too costly, experiment, see what happens, draw lessons, and adapt from there.

Here’s Peter Sims outlining this and other ideas in a pretty interesting 45 minute speech:

On board the Salisbury in 1747

Another great example that sometimes you have to make an experiment to get good data and figure out how things work, comes from the book Adapt by Tim Harford:

Naval surgeon James Lind wanted to find a decent treatment for scurvy, a nasty illness that leads first to spots and gum disease but then to open wounds, internal bleeding, and eventually death. The disease, which still afflicts malnourished people around the world, was then especially common among sailors. Various cures had been proposed. The Admiralty, which commanded the Royal Navy, favoured vinegar.

The Royal College of Physicians took a different view: in its expert opinion, sulphuric acid was just the tonic. Other suggestions included sea water, nutmeg, cider and citrus fruit.

In the spring of 1747, after eight weeks at sea on the warship Salisbury, Lind chose a dozen sailors out of the three dozen then suffering from scurvy. To make his test as fair as he could, he tried to pick men whose illness seemed to be at the same stage. Then he divided them into six pairs and gave each pair a different treatment. The pair being given oranges and lemons made a good recover; those taking cider, acid or brine did not fare so well. It was not a perfect randomized clinical trial by today’s standards, but it did the job. Scurvy, we know, is caused by lack of vitamin C, so oranges and lemons are a sensible treatment.  Ships started to carry greater stores of them, and many sailors on subsequent voyages owed their life to Lind’s experiment.

Lind’s trial highlights, however, some of the difficulties with collecting and reviewing evidence. For a start, if Lind had been tempted to rely on data collected by someone  else for some other purpose – which is quicker and cheaper than organising a bespoke trial – he might have come unstuck. Good data are often just not available: we know from Lind’s account that thirty or forty sailors suffered from scurvy and six men died during that voyage, but official records note only two illnesses. Sometimes there is no choice but to perform an experiment yourself.

Out of Abidjan port

Admittedly, figuring out the details of running a taxi business in Abidjan, seems a bit mundane compared to finding a cure for scurvy, but both endeavors face similar problems in obtaining reliable data.

For example, I tried everything to figure out how much it would cost to get a car out of Abidjan port.  Online there were many stories of problems and issues with getting a car out – including warnings of thefts of parts from cars parked in the port – but no hard official data on prices.  And people who had sent cars to Abidjan reported wildly different  costs, including a top mark of  4 million Cfa franc for a relatively new Mercedes brought in by a European guy with little experience of Africa and no knowledge of French.

The price, which includes getting Ivorian number plates, seems to depend on who you are, your negotiating skills, how expensive the car is, how old the car is, if you are using a “transitaire”, how long time the car stays in port, and well, luck.

I have now gotten my car out of the port, but since it’s  old there’s one more check that needs to be done – probably just a bit of “isomorphic mimicry”.  Anyhow, without the last check (which should be maximum 100,000 Cfa francs) and the payment to the “transitaire” it has so far costed 823,000 Cfa francs which breaks down as follows:

400,000 import duty

250,000 fee for the car being old

155,000 “aconnage”

18,000 storage fee for one day

So it means that in my case, getting the car out of Abidjan port will cost about three times the value of the car (450€) which is well, ridiculous, but I had budgeted with it costing twice the value of the car and it was a budget post I was very unsure of.  Even with this get-out-of-Abidjan-port cost, the figures for the investment still hold up fine.  As long as figures for profits, revenues, and repairs aren’t markably worse than budgeted, and the driver isn’t too dishonest, it should work out.  But that’s to be continued in future posts!

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Categories

%d bloggers like this: