Sunday, 4 February 2018

The Twilight of the Behemoths

The Carillion and Capita stories have highlighted many of the problems that outsourcing gives rise to in the public sector, with notable sub-plots concerning the illusion of risk transfer and the impact of austerity on profit margins, but doubts over outsourcing as a business strategy are also evident in the private sector. The chief argument for business process outsourcing is not direct cost reduction but a focus on "core competence", which is essentially a dilute version of Adam Smith's theory of comparative advantage. In other words, the business should concentrate management on those activities that deliver profit and have the potential to deliver higher levels of productivity. There are two key assumptions here. First, that management is a scarce resource, which in turn justifies high executive pay. But there is little evidence that outsourcing itself intensifies senior management focus. In practice, outsourcing usually means transferring dedicated middle managers outside the corporate boundary. It has little impact on core management. The second assumption is that support services typically suffer from lower productivity due to being relatively labour intensive (i.e. Baumol's Cost Disease). But while many of the functions outsourced may have been characterised by low-productivity growth, this is as likely to have arisen from historic under-investment or poor management as from over-manning.

Sloughing off all non-core activities to outside suppliers makes the firm look lean, but it also means that complexity has been shunted elsewhere; it hasn't been eradicated. Because outsourcers have tended to consolidate, both to achieve economies of scale and lock-in clients by acquiring specialist services, this has led to that complexity and associated risk being concentrated (there is a parallel here with financial derivatives in the lead-up to 2008). Jonathan Lewis, the recently-installed CEO of Capita, has described his business as "too complex" and one that has "underinvested in infrastructure and over-relied on acquisitions for growth". In Coase's theory of the firm, the cost of a business's reduction to its core competence is the increased overhead of supplier contract management. However, that transaction cost analysis rarely considers the implications from the perspective of the outsourcer. Though it has become a cliché that firms such as Carrillion and Capita have become experts in wining contracts, it is less appreciated that this expertise does not reduce the cost of contract management. If anything it exacerbates the problem by ignoring downstream costs. Tales of outsourcers charging silly money for contract amendments may be evidence of price-gouging, but they may also indicate simple inefficiency.

Efficiency gains by outsourcers will have come through a combination of productivity-enhancing technology and lower labour costs, but the ability to make significant improvements through either has declined since the millennium. Much of the growth in outsourcing represented one-off conjunctions in the global economy, such as the first wave of office automation in the 80s, the arrival of the Internet in the 90s (which allowed a lot of customer service to be outsourced to consumers), and the labour-supply shock of globalisation (i.e. offshoring). It is notable that outsourcers are themselves now plagued by under-investment in technology as much as by complexity, while offshoring has lost much of its ability to shave labour costs as global wages converge. The result is that outsourcers now find themselves struggling to achieve significant productivity growth, which undermines their ability to offer falling real costs as contracts come up for renewal, a situation exacerbated in the UK where public sector clients may have no option but to insist on lower prices, all reductions in service scope or quality having been exhausted in the early years of austerity.

If outsourcing appeared to be an irresistible development in the 1990s and early 2000s, the last decade has been marked by an oscillation between outsourcing and insourcing as both private firms and public sector bodies have taken key functions back in-house. While this has sometimes been a temporary measure following an unsatisfactory contract with a particular outsourcer, an increasing number of firms and councils have decided that the initial outsourcing went too far, leading not only to predictable failures in service quality and cost over-runs, but also exposing the organisation to the risks of internal demoralisation and external reputational damage. The Labour Party's current preference for the renationalisation of the railways, and for local authorities to build houses rather than invest in dubious schemes like the Haringey Development Vehicle, echoes a trend that is already well-established in both the private and public sectors. Far from being an ideologically-inspired programme, insourcing is usually advanced for wholly pragmatic reasons: we know it can work (because it previously did so), and we now know that the downsides of outsourcing outweigh the downsides of inhouse provision.

Firms that opt for insourcing frequently cite greater management control as the primary consideration. This is not just control over operating costs or service delivery but also control over data assets. While this is partly driven by "big data" hype, it also reflects a material change in many businesses to a greater reliance on information processing and a wider appreciation of the value of intellectual property, which has caused some firms to question what their core competence actually is. Technology has played a key role in facilitating both outsourcing and insourcing, with the swing to the latter reflecting developments such as cloud computing and software as a service, which obviate the need for up-front capital and allow firms to invest in activities such as data analytics and software development without a comparable investment in infrastructure support. Wider social developments have also played a part with the emergence of low wage regions in developed economies allowing firms to onshore certain functions that in the past might have been offshored, though this also leads to the business being fragmented into geographically dispersed silos, which entails its own transaction costs.

What this highlights is that the traditional division between outsourced and insourced, which was coterminus with functional organisations (e.g. customer service or the IT department), is being blurred. This is another way of saying that the boundary of the firm is now more fluid and dynamic and thus closer to the Coasian ideal. From the outside, it is hard to tell what functions a business operates itself and what it outsources, though this is rarely a matter of concern for its customers. As the private sector has become more discriminating, outsourcing contracts have tended to become smaller, more specialised and more flexible in their management. This has been to the benefit of smaller, more specialised outsourcers, many of which are indistinguishable from independent consultancies, and explains why the larger, traditional outsourcers such as Capita and Serco have been struggling. Scale is no longer an unalloyed benefit and a reliance on big contracts has come to be seen as risky, and not just in respect of public sector clients. The outsource market has become more plural, more varied and more bespoke, which is a sign of maturity.

Though the current headlines are focused on historic errors likes PFI deals in construction or expensive facilities management contracts, what's really going on is the decline of the big "one-stop shops" across both public and private sectors. This same structural change is evident in large-scale housing. The Haringey Development Vehicle was not merely a poorly-designed plan that would have disadvantaged borough residents, it was an out-of-date, 90s-era approach to the problem of supply that sought to create another behemoth with a long-tail of risks for the council. Given that London already has an over-supply of luxury new-builds and is probably facing a crash in prices at the top of the market in the next few years, it is possible that Lendlease, the developer, was starting to get cold feet over the Haringey scheme. The rejection of the HDV is clearly not a nefarious plot by Momentum, but nor is it just a stirring tale of local democracy. It reflects a wider change that will be characterised in the public sector not just by more insourcing, or more plural and selective outsourcing, but above all by a desire to "take back control".


  1. I agree with your take that sentiments on insourcing span public and private domains, against the view that we technologically continue to modularise and overlap in all our activities. To me Coase is a re-boot of the old puffendorf smith controversy about the boundaries between the the family (Oikos//firm) and the 'other tribes' that wandered from the ark into a 'state of nature (market). What marks Coase apart (and so initiated the Chicago research program in the last century) was to declare law itself, as made by human judges, subordinate to price theory., giving us the beloved transactions cost models of analysis. (Thank you dear Ronald). Even the family (patriarchal/hierarchical/vertical) could be analysed as within the law of nature (orphaned, struggle, horizontal)
    For me the failure of these firms at the Political Economy level, is less of interest as an analysis of productively, efficiency, control etc, but more that wider feeling that it's now that time in the book when Oliver stopped asking for more and came home.

  2. Herbie Kills Children5 February 2018 at 17:40

    This always comes down to the bottom line. If outsourcing is cheaper it will happen, if not it won't.

    What usually happens is that people making such decisions listen to the salespeople and it all sounds so wonderful they forget about the pitfalls. And when they discover them it is too late.

    And when I say bottom line it isn't all about outsourcing to organisations that have less worker rights etc, sometimes having your data in the cloud is cheaper than on premise, i.e. economies of scale.

    This is a challenge for a socialist society, how do we get rid of all that and still retain the 'innovation'. Personally i would argue that more cooperation as opposed to competition and more directly targeting problme or goals will improve innovation but it is still a technical issue for a socialist society.