Satellite builders are facing mounting criticism about a falling off in quality in the wake of problems such as those currently faced by Matra Marconi Space (MMS). However, sources within the space insurance industry have suggested that the reduction in quality has been in part at customers’ request.

One underwriter agreed that the number of satellite in-orbit failures had rocketed in recent years. From accounting for about one third of all losses in 1995, last year they had reached 64.82 per cent, only a little sort of double the rate three years earlier. This level should be viewed in light of the fact that the number of satellites launched has increased. Excluding apogee motor and early-orbit failures (the most common cause of post-launch failures), satellite-related troubles account for almost 10 per cent of all losses.

Satellites used to be built with adequate built-in margins, for DC power as for every other parameter; a 3dB power margin (equivalent to 50 per cent) was quite common. This gave manufacturers a 15 per cent loss margin before any question of claims arose. Now, under the "Better, Faster, Cheaper" rules, any parameter which doesn’t come up to the levels demanded automatically results in a "below spec" claim. Effectively, margins are a thing of the past, said the underwriter. "We would have paid out vast amounts if this policy had been operated in the 1980s," he said.

Officially, the problems with the MMS satellites which have not yet been launched are not space insurance concerns; the underwriter interviewed by Interspace normally only initiates cover once the spacecraft leave the manufacturer for the launch site. Before this, product quality – as with every kind of product – is between customer and prime contractor. It is not uncommon for the customer and prime contractor to demand from people further down the chain evidence that adequate cover is in place. Distinctions are ofter made between products that are off-the-shelf, or made bespoke to the customer’ needs. In the past, virtually nothing would come "off the shelf". But the underwriter questioned admitted that its coverage cost would depend largely on its knowledge of the reputation of all suppliers along the chain. The only factor applying specifically to satellites is that minimum power levels specified are generally those available at end-of-life, which nowadays may be 15 years. It is very difficult to test for this.

As the satellite business is much more driven by competition than in the past, CEOs may be becoming indifferent to losses or delays: either the customer pays, or the insurer. The underwriter pointed out that military contracts can cost five times a near-equivalent commercial satellite. Most of this difference lies in redundancy and testing. As commercial satellites become not just bigger but more complex, it is these two which often get discarded first.