On 29 July, Bloomberg New Energy Finance, a Bloomberg-owned consultancy, put out a press release summarizing the preliminary results of a study comparing subsidies to ‘clean’ and to ‘dirty’ energy sources, namely renewable energy and fossil energy. The numbers were quickly picked up and spread, by newspapers, blogs and twitterers, but almost as quickly there emerged in response a growing number of comments and criticisms from the general public and subsidy experts about the reliability of the comparison. What was the cause of their concern? How hard is it to estimate and compare subsidies to different energy sources anyway?

To begin answering these questions, it helps to go back to the exact wording of the Bloomberg press release, which stated that “the world provided approximately US$ 43−46 billion to renewable energy and biofuels technologies, projects and companies in 2009… in stark contrast to the US$ 557 billion spent on subsidizing fossil fuels in 2008, as estimated by the International Energy Agency last month.” Michael Liebrich, CEO of Bloomberg NEF, argued that this should correct the misconception among investors that renewables can only operate if they receive government support, given that “the global direct subsidy for fossil fuels is around ten times the subsidy for renewables” and does not take into account the security and public health costs of fossil energy.

There are a number of problems with these claims even as they stand – to be identified in just a moment – and, as can be expected, the presentation of the information became less nuanced as it made its way across the internet and through the blogosphere. Andrew C. Revkin’s influential New York Times blog, Dot Earth, for example, claimed that fossil fuels get “around 10 times the advantages around the world as non-polluting energy sources.” He concluded that rebalancing this distribution of support could be an important step in promoting clean energy without resorting to laws on greenhouse-gas emissions.

A number of readers and subsidy experts interjected on Revkin’s blog, with the World Resources Institute’s Lee Schipper posting comments, and Earth Track’s Doug Koplow posting a comment and publishing his own blog post about the study. It emerged that not only are there many pitfalls for researchers, journalists and general readers to watch out for when it comes to subsidy estimates, but that these problems are far from uncommon. The following list sums up some of the key arguments that were made in the discussion around the Bloomberg press release and subsequent coverage – how, if it really needs to be done, apples can best be compared with oranges, and the caveats that need to go with any such analysis.

1. Stating estimates in common units: comparing two subsidy estimates as an absolute value will be misleading if different amounts of the thing being subsidised are produced or consumed. Vastly larger quantities of fossil energy are consumed every year than renewable energy, accounting for more than 80% of the world’s total primary energy demand in 2007, according the International Energy Agency’s (IEA)World Energy Outlook 2009. Many commentators pointed out that Bloomberg could have more helpfully compared subsidy estimates expressed in the same units, such as dollars per gigajoule.
    
2. The need for systematic, comprehensive and transparent analysis: Doug Koplow – citing his work reviewing subsidy estimates by the U.S. Energy Information Administration – argued that subsidy accounting needs to be systematic and comprehensive if it is to be valid, and cast doubts that the Bloomberg comparison could have fully met this standard. The press release itself noted that only ‘direct subsidies’ were measured, begging the question ‘how big are the indirect subsidies and would that change these relative estimates?’ This opens up the wider issue that if the results of any study find themselves in the public domain and are influential, it is important that the original research can be accessed and evaluated to assess its robustness. According to Bloomberg NEF, the report is only available to their paying clients, and Subsidy Watch was unable to elicit any more information.
    
3. Different subsidies serve different purposes: the common generalisation is that fossil energy technologies are mature whereas renewables are still in development and therefore deserve higher levels of state support, but this is in fact rather simplistic – many fossil energy subsidies in developing countries are at least ostensibly intended to provide support to the poor, thus serving a very different purpose than their renewable counterparts. Moreover, some ‘renewables’, such as conventional biofuel technologies, are well developed. Although aggregate estimates are certainly valuable, nuance is required to remind readers that total spending, even on a per unit basis, is not the whole story.
    
4. Not all ‘renewables’ are renewable: another common problem with comparing fossil energy with renewables is that it is all too easy to assume that subsidies to renewables are necessarily a ‘good thing’ because – as Revkin claimed – these are “non-polluting energy sources”. In fact, some renewables, including conventional biofuels, which receive a large share of subsidies, can be unsustainable.

Ultimately, the problem of comparing subsidies to different energy types comes down to the fact that subsidies and energy are both complicated issues. It is very difficult to make generalisations that are true across the board, and a range of indicators – taking into account spending, purposes, impacts and opportunity costs – is needed for a full understanding of how spending compares.