The €838m Kosovo highway (which has in recent times attracted the Guardian’s and Foreign Policy’s attention) might well have been one of the most wasteful infrastructure projects in history. Andrea Lorenzo Capussela gives all the numbers: a staggering ‘€400m of the final €838m final construction cost is excess price’, equivalent to 10 per cent of Kosovo’s GDP. ‘Money paid for no reason’, he concludes, from a poor country to a rich foreign firm.
The pyramids hardly improved the welfare of contemporary Egyptian society and must have cost a vast amount of public revenue. Besides their immediate function, however, they probably served also to affirm the authority of the sovereign or to instil a sense of pride in its subjects. This might be true also for Kosovo’s new road. But unlike the pyramids a smooth four-lane 77 km-long grey fenced asphalt strip is unlikely to bring many tourists and foreign reserves to Kosovo four millennia from now, for it lacks their originality – there are other highways – and boldness of design.
It is true that roads are more productive objects than artificial stone hills. But this highway is less useful than it might appear. In 2010, when construction was decided, direct trade between Kosovo and Albania was very low, and Kosovo’s main import and export routes traverse Macedonia and Serbia instead. The same is true today, one year after the highway was completed. Economic integration between the two Albanian nations is certainly likely to grow, but improvements to the existing roads would have sufficed to support its evolution over the medium term, at a fraction of the cost. Indeed the highway is little used, as Foreign Policy remarks, despite still being free of charge.
Kosovo did need investment in the transport sector, but it was irrational to concentrate it on that route because the country still has only 3.6 km of roads per 1,000 residents – the average in Europe and central Asia is 8.6 – and, according to the World Bank (here, pp. 40–42), most rural roads are in ‘bad’ or ‘very bad’ condition, which harms the development of agriculture and adds to the burden of the most disadvantaged segment of the population. A highway linking it to Albania and the high seas does have undeniable geopolitical importance, for Kosovo is landlocked in a still unstable and potentially hostile region: both governments in fact explained their investment as an act of ‘patriotism’, and Kosovo’s citizens have welcomed it. But the country enjoys the explicit and reliable protection of NATO and the West, which greatly reduces the urgency of such geopolitical considerations.
The question was one of timing and sequencing, therefore. A decade or two from now this highway may prove very useful. But in 2010 Kosovo suffered levels of poverty, unemployment and infant mortality that ought to have imposed other priorities instead, such as education, health, water supply, energy supply, or indeed other parts of the transport sector. If this project had a rationale, it probably resided in the manner in which it was executed.
The procurement process was radically flawed, because bids could not be compared according to objective criteria – crucially, both variable and fixed prices were accepted – and the terms of the construction contract were negotiated after the winning bidder had been selected, when the negotiating power of the government was lowest. The tender was won by a consortium led by a large US company, with a price estimate of about €400 million. But during the subsequent negotiations the price rose by more than 60 per cent (to €659 millions). Donors had provided the government with a specialized legal adviser, who warned it that the terms proposed by the consortium were ‘not compliant’ with the tender rules, and were ‘extremely dangerous’ because ‘the apparent contract price [was] nothing more than a non-binding estimate’, liable to rise during construction through revisions, penalties and extra costs.
The seriousness of this risk was illustrated by a close precedent. In 2006 the same consortium had won the tender for the mountainous segment of the Albanian tract of the highway, with a very similar variable-price contract: the initial estimate was €418 million, but the final price reached €1 billion. Nonetheless, the government accepted the terms proposed by the consortium, with some superficial changes. And although a few weeks later it acknowledged to the IMF that the contract ‘may not adequately protect the budget from cost overruns’ (p. 38 of this IMF report), the government allowed construction to proceed for more than a year without supervision, which is a crucial safeguard – contemplated by the contract itself – precisely against the risk of unwarranted cost inflation.
The government thus made a succession of manifestly irrational choices, each of which significantly increased that risk. Supported by the IMF, the European Commission and the World Bank, the international supervisor of Kosovo – the International Civilian Office (ICO), whose economics unit I then led – had attempted to avert them. But eventually it desisted, effectively preventing anyone else from intervening.
A year later, asked by an interviewer why the ICO remained silent, the head of mission answered, ‘ask the US’. He did not elaborate further*, but both the Guardian and Foreign Policy have explored this angle and reached the following findings, which correspond entirely to my impressions: the then US ambassador supported this project, ‘lobbied’ for the consortium, and even ‘put pressure’ on Kosovo’s government to accept those ‘extremely dangerous’ (according to the government’s legal adviser) contractual terms. Shortly after leaving Kosovo, moreover, this person was hired by the US firm leading the consortium. Whatever the motives, at any rate, we may proceed upon the assumption that Washington’s representative – the most powerful diplomat in Kosovo – supported this project, whose cost exceeded the initial estimate.
The government’s acknowledgment of a possible risk of ‘cost overruns’ proved to have been a heroic understatement, in fact. Even though the length of the highway was significantly shortened, the construction price rose by a factor of 2.7: the bid by which the consortium had won the tender was €400 million (for 102 km), or €3.9 million per km; the final price was €838 million (for 77.4 km), or €10.8 million per km (this was only the price paid to the consortium: the overall cost, including expropriation and other peripheral costs, reached €1.13 billion). The final per-km construction price is between 40 and 50 per cent higher than comparable EU averages (calculated by the European Court of Auditors) and 2.5 higher than in Germany, and it seems grossly excessive for Kosovo: for instance, workers employed by the consortium, or its subcontractors, were reportedly paid only €1.35 per hour.
Let’s assume that the initial estimate (€3.9 per km) was a fair price (it seems rather high, conversely, as it is only marginally lower than average German prices, where cost levels are much higher than in Kosovo). And let’s also assume that one third of the subsequent price inflation (2.7 overall) was objectively justified. Based on these rather favourable assumptions, we can safely conclude that at least €400 million of the €838 million final construction cost is excess price: money paid for no reason.
To appreciate the scale of the problem let’s express these numbers as percentages of Kosovo’s GDP in the year when construction was decided (see the latest IMF report on Kosovo, at p. 18). The amount of public money that was misspent – i.e., the total cost (€1.13 billion) of this (for now) largely useless object – is equivalent to about 26 per cent of the 2010 GDP. The amount of money that was wasted – i.e., the €400 million excess price – is equivalent to about 10 per cent of GDP. So the pyramid parallel (parallepyramid) was perhaps less of a caricature than it might have seemed. By way of comparison, in fact, in Italy 10 per cent of GDP is equivalent to about €140 billion: and with 26 per cent of it one can probably build a second Sicily, with Greek temples, Abbasid mosques, Norman castles, baroque palaces and the rest.
Such disproportions are predominantly to be ascribed to the mistakes committed by the government in the procurement, negotiation and execution phases. And as each of such mistakes were made consciously, the most plausible explanation for them is the abuse of public office: corruption, presumably, which is likely to have influenced the very decision to undertake this project. Our interpretation must remain conjecture, however, because no direct evidence of corruption has emerged and the EU rule-of-law mission in Kosovo, Eulex, has not even opened an investigation on these matters, despite having received (from me: see § 2.1 of this paper) documents which prove that the government knew that its choices were contrary to the public interest.
To conclude, this highway illustrates very well some of the reasons why the state-building intervention in Kosovo largely failed, as I argue in a book that shall be published next month: weak legal and political accountability (the root cause of the government’s mistakes); divisions and confusion within the international community (those wielding informal power prevailed over those entrusted with formal ones); the weakness of Kosovo’s supervisors (both abdicated from their duties). So the EU and its member states, which have an interest in the long-term development of Kosovo, had to watch powerlessly as its government imposed a huge opportunity cost – namely, the more useful objects (schools, hospitals, rural roads, water pipes, etc.) that could have been built with 26 per cent of GDP – on a still poor and rather fragile country. Indeed, the most remarkable aspect of this story is that it happened in the world’s most closely supervised nation.
One last comment. Few rich people from rich countries would, if asked, refuse a €220 donation to a developing country. But few citizens of a developing country would donate any money to a rich foreigner. In Kosovo they are more generous than that: each of them made a €220 gift – the €400 million excess price divided by Kosovo’s 1.8 million resident population – to the shareholders of the consortium; and as its dominant member (Bechtel) is a privately held corporation we can assume its shareholders to be rather few and rather rich.
The gift was all the more generous on the part of the Kosovars because the recipients were far less likely to spend it in their towns (Pristina, Ferizaj, Brezovica) than elsewhere (Manhattan, Urbino, Sankt Moritz). So it is to be hoped that at least a small portion of the excess price was transferred back to Kosovo’s public officials, for they are quite likely to spend or even invest some of it where they live, thus stimulating economic development there.
* Both articles quote the head of mission (Pieter Feith) as saying that he could not intervene because the ICO lacked, or was denied, the necessary information. My recollection is different: well before the contract was signed I sent Feith the most relevant documents and a note outlining the main problems; the note bears his handwritten comments, and was the basis for a few conversations between the two of us.
Note: This article gives the views of the author, and not the position of LSEE Research on SEE, nor of the London School of Economics.
Andrea Lorenzo Capussela has a PhD on competition policy and is a lawyer by training. After a few years in the private sector, he served as the head of the economics unit of Kosovo’s international supervisor, the International Civilian Office, in 2008–11, and as the adviser to Moldova’s economy minister and deputy prime minister, on behalf of the EU. His forthcoming book State-building in Kosovo: Democracy, EU Interests and US Influence in the Balkans, I.B. Tauris, London, will be presented at LSEE in May 2015.