Horse meat scandal: the long face of toxic regulations

European consumers were alerted to their ignorance by the horse meat scandal. But this isn’t the first time complex systems have failed

For a nation that loves horses, it came as something of a shock when the Food Standards Agency in the UK discovered some ‘beef’ products – such as frozen lasagnes and burgers – were more horse than cow. Of course, the scandal posed a risk more to cultural prejudices than to the health of the population. However, it is a reminder that modern supply chains offer society a high level of efficiency, with the drawback that we often have little clue what is going into the things we consume.

Such supply chains are often tortuously complex: here it involved a French-owned factory in Luxembourg, another French company called Spanghero, a Cypriot trader, another trader in the Netherlands, and finally an abattoir in Romania (the Romanians have since denied the mislabelling originated with them).

Not that long ago, consumers would obtain their minced meat by going to the butchers and asking for a specific piece of meat to be ground up. It would have been difficult to end up, by accident, with a piece of a different animal species from an East European abattoir.

But today, rather than relying on our own eyes, we have to trust a hugely complex international supply chain. We believe (or accept) that the label on a product is accurate, and the contents have not been changed or substituted at any point in the chain, through error or fraud.

Meat mixing
Food has many powerful cultural and emotional resonances, which is why the horse scandal got so much press. The same kind of meat mixing occurs in other areas, such as financial services. Consider, for example, the changes in the US mortgage industry that helped kick off the financial crisis.

In earlier, simpler times (like the 1980s), mortgages were traditionally a straightforward bond between the homeowner and a financial institution. A disadvantage of this arrangement was that the bonds were hard to trade, so banks had to keep the low-yielding liabilities on their books for a fixed term. To address this problem, in the 1990s, they turned increasingly to methods such as collateralised debt obligations (CDOs).

Internal S&P documents show that the company regularly tweaked its models to give the right (i.e. triple-A) answers – which was nice for their clients, but less so for the purchaser

CDOs consist of bundles of mortgages, which can then be divided into tranches of varying quality and sold as separate instruments. On the surface, they offered a neat way to take a group of individual mortgages with different default risks and payment schedules, and homogenise them into an easily traded, plastic-wrapped product with a tailored degree of risk.

The mortgage ‘supply chain’ became increasingly complex and specialised: a broker would sell mortgages to homeowners; these mortgages were then compiled by a mortgage bank; an investment bank would transform them into an investment product; another firm would be responsible for managing payment collection; and a rating agency would stamp the whole thing as grade A prime beef.

As with the food supply chain, this complexity led to some cost efficiencies, but also had the effect of severing the connection between mortgage supplier and homeowner – so their interests were no longer aligned. The processing and packaging also meant that the end purchaser had little idea what went into it, which left the system open to abuse. The result was far more toxic than any equine lasagne.

A financial mess up
The people in charge of labelling this financial equivalent of mystery meat were the rating agencies, such as Standard and Poor’s (S&P). The labelling acted as a reassurance that the product was safe. Unfortunately, these institutions didnít do a very good job. The products again often turned out to contain more horse than beef.

Part of the problem was the mathematical models that were being used by most agencies. These were based on historical records of volatility. Since the US housing market had been growing steadily for decades, it is no surprise that they underestimated the risk of default.

The rating agencies also faced an uncomfortable conflict of interest. In order for instruments such as CDOs to be attractive to investors, they needed to have a high rating, preferably triple-A. When the housing market started to tank in early 2007 (if not before), these ratings should have been adjusted down: but this would not have pleased clients, such as major banks.

So while the agencies were in theory objective and independent, they actually had a stake in keeping the ratings high in order to allow their clients to get the bonds off their books. According to a recent US civil complaint, internal S&P documents show that the company regularly tweaked its models to give the right (i.e. triple-A) answers – which was nice for their clients, but less so for the purchaser.

As Tony West from the US Justice Department put it: “It’s sort of like buying sausage from your favourite butcher, and he assures you the sausage was made fresh that morning and is safe. What he doesnít tell you is that it was made with meat he knows is rotten and plans to throw out later that night.”

Systems theory
Globalisation in industries such as finance or food has been hugely beneficial in many respects. It is indeed a miracle that human beings spread over many different countries can conspire to make single products in such an efficient manner.

But complex systems theory teaches us that efficiency often comes at the expense of robustness. A system that is highly efficient, and minimally regulated, may also be sensitive to small perturbations, or vulnerable to contagion. For this reason, many biological systems, such as those found in the human body, are far from paragons of efficiency. The regulatory bodies that provide a degree of oversight might make products more expensive – but as we learnt at the supermarket, you get what you pay for. And sometimes, it really is better to buy local.

Related topics: