The Murphmeister tries to understand insurance

With predictable consequences.

Last year, Glencore UK\’s derivative trading with other parts of the group totalled $383bn (£267bn), more than twice the yearly budget of the National Health Service. The practice resulted in a $122.8m loss for the London-based business, effectively docking that amount from UK profits and transferring it to the main group based in the low-tax Swiss canton of Zug.

Richard Murphy, of Tax Research UK, said: \”Glencore is insuring itself with itself. If I insure my house for fire with myself and it burns down, I\’ve got to pay myself for the house which has burnt down. That\’s what Glencore is doing, and the consequences are that the risk is never leaving Glencore; it\’s still inside the group. That\’s $383bn worth of trades that, on the face of it, make no sense whatsoever. We don\’t know, but it is highly likely that the motivation is not genuine insurance and it looks like a significant amount of tax planning takes place within this trading function.\”


It\’s entirely standard that when an organisation becomes large enough it becomes self-insuring.

I have no idea at all how WalMart insures itself so let us take them as a theoretical example.

If you\’ve one shop on which your entire livelihood depends then it makes sense for you to insure that building with an outside insurer. They take the risk, not you, of a low probability but high expense event.

They, of course, are insuring many such singly owned shops and can do the risk spreading of collecting all the premiums and paying out on the few fires.

Excellent, risk is spread and everyone is happier.

Now, move to being WalMart. You\’ve got umpty bazillion shops right around the world. You shuffling the fire risk off onto an outside insurer doesn\’t really make any sense. You\’ve already got a geographically spread risk, you\’ve already got enough stores that one burning down isn\’t a low risk but high cost event. You\’re probably actually at the point where one or another of them burning down is a certainty in any one year. Why let someone else wax fat off the investment earnings from the premiums?

Sure, you want all of the individual stores and units to act as if there is insurance, but risk to the organisation overall should perhaps be kept inhouse. So, you have an inhouse insurance subsidiary. Note that I\’ve no idea whether WalMart does: but I know that many oil companies do.

And this is what Glencore is doing. Self-insuring because they\’re a large enough organisation to do that.

Oh, and guess what? There\’s another reason the Murph is wrong as well.

In 2010, Glencore UK made a profit $186.5m on insuring itself with its parent,

They are actually insuring themselves too. This year\’s numbers are just part of the swings and roundabouts of such insurance.

10 thoughts on “The Murphmeister tries to understand insurance”

  1. Murphy doesn’t economic, tax, common sense, how the hell is someone of his “unique” intellectual skills ever going to understand derivatives.

    $383bn worth of trades, “transferred” $122.8m worth of profit. If that’s the best Glencore can manage, they are pretty pathetic. Looks like the most complicated tax management strategy ever, for the smallest benefit.

  2. Same with very expensive aircraft, above a certain fleet size depending on replacement cost there is no point in buying hull insurance.

  3. No, you don’t understand the Murphy perspective. How dare the evil corporatist looters transfer profits out of the UK under the pretext of insurance? That money belongs to the UK Government! They have stolen it from us!

  4. So in 2011 net insurance payments out of Glencore UK to the parent were $122.8m.

    In 2010 net insurance payments into Glencore UK from the parent were $186.5m.

    So over two years, this process has increased UK taxable profits by $63.7m.

    If that’s Murphy’s idea of tax planning, no wonder he left private practice.

  5. It is more common practice is to have an in-house *reinsurance* subsidiary as various governments require companies to take out insurance through an authorised insurer for certain risks (employer’s liability, motor third party etc) but the company can accept the risk through a reinsurance contract with the primary insurer.

  6. There is also such a thing as catastrophe risk. Suppose that we expect 10 events a year of average cost. The insurance premium might be, say, the cost of 12 events a year. Best self insure. But maybe worth insuring against a cost value of, say, more than 15 events. The likelihood is very small, so premium correspondingly low.

    Quite what is best, will depend on the shape of the distribution curve of the cost of insurable events.

  7. Bit confused about terminology here. Murphy complains about “derivatives trading”. But isn’t trading in the swap/hedge, er….insurance created by an ISDA a qualitatively different action from the creation of the ISDA? the latter’s the insurance, the former’s an attempt to buy/sell an attractive contract….

Leave a Reply

Your email address will not be published. Required fields are marked *