Thursday, July 16, 2009

Time to bury the Oxford report

On several occasions we have written about a report produced by the Oxford Centre for Business Taxation, which is critical of estimates of illicit flows and other offshore-related phenomena published by TJN and its colleagues. (Update on comments here.)

Professor Michael Devereux of the Centre has replied in the Financial Times to an earlier letter from TJN and its partners. Devereux’s riposte says little of interest, but accuses us of seeking “to spread innuendo about the messenger rather than to engage in constructive debate about the research” – without noting that the letters page of a newspaper – ours was just 263 words long – simply isn’t the place to offer a detailed rebuttal of a detailed report.

A blog, however, is an excellent place to do this. So here we go!

We’ve already exposed some of the fatal flaws in the report in several earlier blogs; this blog seeks to look at this report in a slightly more organised way. In short the Oxford report cannot be supported by their research, and the researchers are not in control of their materials. The estimates by TJN and its colleagues are by far the best estimates that are out there, and the report has strengthened our case for further research in this area, (and we only wish that the World Bank and others would get on the case as we have been urging for so long). The Oxford report has quite simply failed to knock our numbers down -- and what doesn’t kill us, as we recently remarked, makes us stronger. They haven’t, it should also be noted from the outset, produced any numbers of their own, though that wasn’t the focus of their report.

Until more research has been done, ours remain the key references in this area.

This long blog is in four sections:

1. Errors of analysis
2. Errors of omission
3. Errors of fact
4. Process, Context, questions and conflicts of interest

Errors of Analysis

There are too many errors of analysis to cover exhaustively here. We will focus on just a few important ones. One key mistake – reproduced in the Financial Times as a potential headline rebuttal of TJN’s and other estimates, goes like this:

“Tax revenue losses due to mispricing are overestimated drastically.”

This is unsupported and, as we shall see, the researchers must have known that this was unsupported. To appreciate this, it is necessary to look at the paragraph in which the sentence is embedded:

“A key shortcoming of many existing studies based on mispricing is that they only take into account overpriced imports into developing countries and underpriced exports of these countries. But the mispricing approach would also allow to identify underpriced imports into developing countries and overpriced exports. Both shift income into developing countries. Estimates of tax revenue calculations have to take into account income shifting in both directions. If only one direction is taken into account, the results are misleading. In this case, tax revenue losses due to mispricing are overestimated drastically.”

That last sentence is of course, the headline from this entire report. But it is just plain wrong. Here’s why.

At first glance, they might seem to have a point. For completeness’ sake it would certainly be useful to measure the flows going in both directions, as they say. However, let’s consider what this means.

What TJN’s colleagues have measured in terms of mispricng is capital flight out of developing countries into secrecy jurisdictions and other locations. What it does not measure is capital flight into developing countries. To illustrate: TJN’s colleagues have measured capital flight out of, say, Congo, and into Switzerland – but have not measured capital flight from Switzerland into Congo. This is not to say that there is not an issue here – there is – but this now brings us to two important points.

First, it is fanciful to think that capital flight into developing countries from tax havens and elsewhere is likely to be anything like as big as capital flight out of developing countries. Even if, say, capital flight into developed countries were, say, 20% of capital flight losses out of developing countries – which seems unlikely (see Richard Murphy’s analysis of this here) - and we were to subtract one from the other, that would be no grounds at all for concluding that there has been a “drastic” overestimation.

In addition to this, how could they possibly come up with “drastically” if they have not measured this themselves – which they haven’t? The use of this -- headline soundbite – word cannot be supported.

Yet the second point is more fundamental. For the researchers have made an elementary yet crucial error. They say we should subtract tax revenue losses in one direction from tax revenue losses in the other direction – when in fact we should add them.

If a country loses tax revenue from overpriced imports into developing countries and underpriced exports, it does not somehow magically recoup illicit money going in the other direction which suffers losses from, say, flows evading VAT or import duties. No, it loses revenue in that direction too. Dev Kar, a former Senior Economist at the IMF who put together the data for Global Financial Integrity, has sent us a more detailed analysis of this particular issue. His full analysis is here (and NB he has now added a useful further comment at the bottom of this post) but the key section says this:

“The traditional approach to estimating annual illicit outflows through trade mispricing has been to simply take all signs as they are and allow them to wash out into a net position. For instance, if the CED (Change in External Debt) model estimates illicit outflows of 100 but the net trade mispricing (based on Direction of Trade Statistics or the DOTS model) shows inflows of 150 (i.e., -150), then, according to the traditional method, the country has received a net inflow of 50. No economists have questioned this “traditional approach”
. . .
the traditional method yields strange regional distribution of illicit flows—Africa as a whole receives illicit capital (which flies in the face of its continued dependence on bilateral aid). The traditional method would also have us believe that Russia is a net receiver of illicit capital." (Once again, read the full analysis here.)

Let’s repeat that last sentence. No economists have questioned this traditional approach. Perhaps it is time for a certain section of the economics profession to make some downwards journeys from their ivory towers, all the way down to street level.

Or, as Murphy put it: the inward and outward abuses do not match in terms of transactions, and do not net out against each other. “No one is going to do dual transfer pricing abuse in and out to arrive at a net correct position.”

From this one can only conclude that our estimates, by only considering flows in one direction, must be underestimates, not overestimates, of tax revenue losses.

What were the Oxford researchers thinking?

Next, the Tax Justice Network’s analysis is not, as the researchers put it, a “rough back of the envelope calculation.” It is based on data drawn from the Boston Consulting Group, Merrill Lynch, and Ernst & Young Cap Gemini, with supporting evidence from the Bank for International Settlements. They say our cited rates of return are based on “strong assumptions”. Yet our rates are based on considerable research at the practitioner coalface, and are put together by a practicing and highly experienced chartered accountant and a former Economic Adviser to the British tax haven island of Jersey. We cannot allow academics to dismiss these as “strong” without any basis for dismissal. They entirely misconstrue the research by TJN and others. Read more on this, and why our estimates contained some extremely conservative assumptions, here.

Errors of Omission

The Oxford report is entitled “Tax evasion, tax avoidance and tax expenditures in developing countries: A review of the existing literature.”

A literature review should aim to take a look at all the best estimates out there. But why has the Oxford report been so selective in this review? We cannot answer that for them. We can only conclude that either the researchers were incompetent and did not know about these reports, or they purposefully excluded results. The former is extremely unlikely, given that the different estimates, and many more are all laid out and clearly linked to, here. They miss, or avoid, crucial sets of estimates such as:
  • Offshore Explorations: the most detailed and thorough study of some of the world’s most important secrecy jurisdictions, by the high-brow tax publication TaxAnalysts.
  • Research by James Boyce and Léonce Ndikumana of the University of Massachusets, Amherst, estimating capital flight from 40 sub-Saharan African countries from 1970-2004 at $607 billion in 2004 dollars (including interest earnings), compared to $227 billion external debt in 2004.
  • Assessing the Development Squeeze: Income Loss to Less Developed Countries as a Result of Tax Incentives, Tax Evasion and Tax Avoidance (2004) Alan Muller, David Frans and Rob van Tulder, SCOPE Expert Centre, Potterdam School of Management.
  • Refinements in the work of Alex Cobham in the 2007 report prepared for the Oxford Council on Good Governance.
Why did they not mention estimates such as these in a literature review? We will take this question up again in the section under Process, below.

They have also omitted to define their terms. What is a tax haven, or a secrecy jurisdiction? We mention this partly because a paper published by the Oxford Centre For (or should we say Against?) Business Taxation seems to think that tax havens are all about tax, when we know very well that the more fundamental issue is secrecy. That is why we use the terms “tax haven” and “secrecy jurisdiction” interchangeably.

Finally, there is another important error of omission. They did not produce their own estimates to back up their case. They simply sought to knock down what was published. Now in a sense that is fine – this was a literature review, after all, and their recommendation that “more research is needed” is of course well received – but the ways that they set about filling the holes raise serious questions about their experience and understanding of this field. Here are a few examples.

1. The report suggests a future programme of study that would require accurate assessment of the role of tax havens based on firm-level data. All we can say is this: “Good luck, guys!” Do they really think this is achievable? It would be virtually impossible to do this, due to the secrecy veil offered by many havens: such data is impossible to secure. Do these researchers have any understanding of these issues? As our letter in the FT stated: “It is a shame that such prominence was given to a first contribution from researchers whose expertise in corporate finance, rather than development, has perhaps hindered their ability to consider problems where data are inevitably limited.”

2. A “micro” approach that they suggest for estimating the tax gap based on tax audit data of individuals or firms would prove to be impossible in many developing countries due to the ease with which full information on income, particularly offshore income, can be hidden. Strangely, they add this: “Data to implement micro methods is seldom available for developing countries.” What are they saying – use this data even though it isn’t available? Once again, good luck guys on getting the data. Read more here.

3. Estimating the tax gap based on national accounting data, as in the Swedish example, would be impossible in many developing countries due to inadequacies in the collection and compiling of accurate data on sources and uses of funds. It may be that the IMF and the World Bank could come up with something useful here, at least in a rough and ready way. But they haven’t. Why not? Are developing countries not a priority?

And we should refer to one other sentence in our letter to the FT:

“It is not reasonable to treat absence of the data needed to conduct strictly conventional analysis as evidence of absence of a problem. Instead, this implies that more transparency is needed.”

To suggest that this problem cannot be proven to exist because it does not fit the modelling capacity of conventional economics does not suggest there is nothing wrong with the world: it suggests that the modelling techniques of economists must be developed in ways we and our associates have pioneered.

Once again we have pointed out how so many economists at the IMF, World Bank and elsewhere have conflated an absence of data with the absence of a problem. They haven’t measured this stuff – and they still aren’t measuring it (see more here.) For this reason, the Oxford report which concludes that this research is required, is to be welcomed despite its shortcomings. This is almost certainly the biggest data gap in the entire global economy.

(There are further errors of omission in the fourth section under Process, Context, questions and conflicts of interest.)

Errors of fact

There are a number of factual errors worth pointing to. Here is a selection:
  • The authors completely misstate the basis of the study of illicit financial flows by Global Financial Integrity (GFI), which draws on extensive data sets for IMF Direction of Trade Statistics and for the World Bank Residual Method of analysis.
  • Contrary to what the researchers say, GFI’s methodology is not dependent on categories of items traded or on the quality of goods traded. To suggest that this is so is a gross factual misstatement.
  • Price differences within product groups of the Harmonized Code, as analyzed by Simon Pak and John Zdanowicz, have prevailed in U.S. courts. The argument that such price differences may only reflect quality differences has been legally overruled.
  • Reference to the internationally used Harmonized Commodity Description and Coding System is omitted. This system breaks down commodity descriptions to as many as 25,000 categories and is widely used by almost all governments in trade classifications and customs collection.
  • The database used by Pak is drawn from U.S. sources and is not drawn from the customs agencies of other countries.
  • Apart from the glaring error pointed to above, GFI’s analysis of illicit financial outflows from developing countries is conservative, since it does not include a number of potentially very large elements that were simply not included in the measurements. These include
    - Trade mispricing that occurs within the same invoice
    - Illicit financial flows due to smuggling and trafficking
    - The mispricing of services.
  • The notion inherent in the report that “illegal activities which would be stopped if detected and thus would not generate tax revenue” is wrong. To cite a few examples, oil bunkering out of Nigeria, diamond smuggling out of Sierra Leone, and illegally cut timber exported from Indonesia would indeed contribute tax revenues if shifted into the legal economy.
  • The four permutations of trade mispricing need to be clarified in the report. These are i) between unrelated parties within the same invoice, ii) between unrelated parties via reinvoicing, ii) between related parties within the same invoice, and iv) between related parties via reinvoicing. This is not recognised in the report at present.
  • The Oxford researchers claim that TJN, in its Price of Offshore, has “combined” its estimates from commercial operators such as Merrill Lynch with those from the Bank for International Settlements. Not so. The results have not been “combined” but have been triangulated – we took three quite separate sources of data to see to what extent they converged on a data range. It is a perfectly valid statistical technique. We now know that one element of our analysis – $2 trillion in “other assets” for which data is almost impossible to come by – was drastically underestimated. Read more here.
Process, context, questions and conflicts of interest

The errors of process the academic researchers have committed are strange, but important. The questions about the competence, and more importantly the intentions, of the researchers, simply cannot be ignored. Devereux seeks to dismiss this in his letter to the FT – he claims that the fact his centre was funded at the outset by the Hundred Group of companies is irrelevant. No influence? It happens all the time: read more here.

Here are some specific things to consider.

First – and this is peculiar – the researchers seem to have decided not to speak to or otherwise contact any of the people who worked in this area. Though there is a caveat -- Richard Murphy adds this:

“Of the eleven authors whose work is reviewed by the Oxford team nine are personally known to me. Only one was given any opportunity to discuss his work, and that because he asked to do so.”

And he cites further examples of their peculiar omission in a blog entitled “Why didn’t you ask, Oxford?” here.

This omission could go a long way towards explaining their multiple, often elemental and elementary, errors. “How did you arrive at this?” they might have asked. They didn’t. And they had ample opportunity – TJN’s John Christensen and Oxford’s Michael Devereux debated publicly at the Oxford Union in May, for example (and guess who won the debate hands down – you should have been there!) But at no point did Devereux ask any questions on this crucial research.

“I’ve been a professional researcher for most of my life,” says Christensen, “and normally you begin by talking to people.” Why would the Oxford researchers avoid talking to the authors of the estimates they were studying, even when standing in front of them?

Very odd.

Next, the report was supplied to the Financial Times by somebody before it was published, resulting in the wholly erroneous “drastically overestimated” soundbite being put out there in the public domain. We’re not against journalists getting scoops and digging out hidden documents – far from it – but those who lose control over their documents tend to get shirty about it. What we are doing in fact is asking not so much who leaked this to the FT, interesting as that might be (was it HM Treasury? – note the context of the story where it appears) but more importantly, why?

Next, all of us have repeatedly made the case for others such as the World Bank or the IMF to produce estimates of their own as to the scale of these problems. That has been one of our declared aims in producing our own estimates – to prompt others to start measuring. Yet at no point do the researchers mention this point. Why not? Read more here.

Yet there is something else. The Oxford Centre for Business Taxation does not look like an independent research institute. It looks more like a campaigning organisation or lobby group, which uses a lot of its own research to support its case against taxation of businesses. This is a problem almost generic in academic life – see here. This is by no means a small matter. Funding on international taxation around the world has come in very large measure from corporations. And guess what results have emerged to date? Fortunately, the balance is now starting to shift back.

But we can be more specific than that, in the case for the Oxford Centre for (against) Business Taxation. Let’s start with Clemens Fuest of Oxford, one of the authors of the report. He is, according to his biography, “member of the Academic Advisory Board of Ernst and Young AG, Germany.” Read Richard Murphy’s description of the conflict of interest here.

Yet there is something larger. Take a look at this article in Accountancy Age, reporting on the centre’s aims from the outset:

“The culmination of this mission, he (Christopher Wales of Goldman Sachs) says, was the creation and launch of the Oxford University centre for business taxation (see box) on 4 November this year. Based at the Saïd Business school and backed by £5m-worth of funding from the influential Hundred Group of Finance Directors, the centre has been set the goal of using academic weight, alongside HM Revenue & Customs and business expertise and assistance, to achieve a more competitive tax system for British businesses.”

Two things. First, the last sentence makes it clear they are aligned with British business interests -- or rather, their view of 'British business' which seems to elevate UK-listed multinational companies above all others. (Most British business, providing much needed employment in these difficult times, are smaller players without access to international tax minimising strategies – there is no level playing field.) Second, anyone who is familiar with TJN’s blog and its website knows exactly what “competitive” means in this context (if you don’t – then click here, and read the rest of the Accountancy Age article which shows how we have interpreted it correctly.)

The fact that the centre has this aim means it is more of a lobby group than an independent research institute (we are an advocacy organisation – but we are completely open about that and we seek to serve the public interest, not a narrow set of interests.) The fact that the Centre has no background in development policy, combined with its focus on making UK tax policy “competitive” - raises serious questions about Devereux' claim to be 'independent'.


Blogger Dev Kar said...

There are significant problems with the argument extended by Fuest and Riedel that capital flight reversals need to be taken into account to arrive at a net position with regard to capital flight.

1. A net position can only be taken if the net amounts reflects a net benefit to a developing country. For example, if there has been capital flight out of say Congo of 100 and it received 50 in "reversals", there is a net capital flight of only 50. According to Fuest and Riedel, our approach would overstate capital flight. But they lose sight of the fact that the so-called reversals are also illicit. If the reversals are illicit as well, the government cannot tax them by definition. You cannot tax an amount that does not even show up on any books.

2. Illicit inflows tend to drive the growth of the underground economy and not the official economy. If reversals are unrecorded and illicit, the chances for the official economy benefiting from them are slim.

3. No economic research has validated capital flight reversals much less studied their nature and composition. Illicit outflows can be validated by looking at the growth of deposits in the points of absorption such as offshore financial centers, tax havens, and onshore banks as well as the growth of the underground economy in developing countries. But inflows have not been validated thus far. The traditional approach of automatically washing out flows in both directions does not reflect this simple fact.

7:20 am  
Blogger Clemens Fuest said...

We would like to respond to the two preceding entries in this blog because we disagree with the arguments put forward and we think that it is easy to see why these arguments are wrong.
The main issue seems to be whether it makes sense to take into account what is measured as income shifting out of developing countries and to ignore what the same statistical approach would measure as income shifting into these countries. We restrict our response to this issue (although the other arguments raised in the first entry are equally misguided). In our report, we have made the following point:

The restriction to one direction of income shifting leads to misleading results if the findings are used to estimate the impact of income shifting on corporate income tax revenue collected, as e.g. in recent studies by Christian Aid (2008, 2009), cited in our report. A meaningful estimate of the tax revenue effects would have to take into account shifting in both directions.

It is helpful to demonstrate this using the following simple example: Assume that there are three exporters of a good in country A. Assume all firms have costs of 4 in country A which are deductible from the profit tax base in country A. Firm 1 exports the good at a price of 4, firm 2 exports the good at a price of 8 and firm 3 exports the good at a price of 12. The mispricing approach would identify the transaction at a price of 4 as underpriced and the transaction at a price of 12 as overpriced. The goods are exported to country B, where all three are sold at a price of 14 to consumers in country B.

In this example, the aggregate corporate income tax base in country A is equal to 12. Firm 1 shifts income out of country A and firm 3 shifts income into country A. In the absence of trade mispricing, the tax base in A would be the same. The tax revenue loss of country A due to mispricing is equal to zero. A method which only takes into account firm 1 and neglects the implications of mispricing by firm 3 is clearly misleading. The same applies to the impact of income shifting on country B.

What happens if other taxes are considered? For simplicity, consider the case of tariff revenue losses due to avoidance or evasion (the VAT example used in the first entry is misleading because one would have to take into account VAT paid by the importer; VAT cannot be saved by underpricing imports if the goods are imported and sold on to consumers, at a given price, by a firm which is subject to VAT itself). So assume that there is a proportional import tariff of 10% in country B. Given the prices 4, 8 and 12, the overall import value is 24, and tariff revenue would be 2.4. Now assume that mispricing disappears, i.e. firms 1 and 3 adjust their prices to 8. In this case, import tariff revenue would be 2.4 as well. This example shows that it is of key importance to account for both under- and overpriced imports to B.

How do other economists analyse these issues? The US economist Kimberly Clausing has analysed income shifting through trade price manipulation into and out of the US (Clausing, Kimberly (2003), Tax-motivated transfer pricing and US intrafirm trade prices, Journal of Public Economics 87, 2207-2223.) Her approach differs in many ways from the mispricing approach we have criticized in our report. She identifies mispricing by comparing trade transactions within multinational firms to trade transactions between independent firms. This work has been published in a peer reviewed academic journal, which has a very high reputation.

Clausing’s approach allows for the possibility of income shifting out of the U.S. as well as into the U.S.. She finds that multinational firms manipulate prices to shift income from countries with high tax rates to countries with low tax rates. This may seem obvious, but it is inconsistent with the approach we criticise, and which this blog defends. That approach simply assumes that all profit shifting is from developing countries to developed countries, irrespective of tax rates.

Clemens Fuest and Nadine Riedel

10:10 am  
Anonymous Richard Murphy said...

Clemens Fuest has tried to justify his belief that tax driven inflows to developing countries are as important as tax driven outflows. To do so he has offered a purely hypothetical model which he says justifies his claim.

Such may be the approach of the economic theorist. The work we do deals with economic realities.

For his claim to be a plausible conclusion in the world of economic reality there would have to be a body of evidence suggesting that:

a) There is prima facie evidence that people want to shift money into developing countries through illicit financial flows based on transfer mispricing;
b) That there is evidence of them actually doing this;
c) That there is documented evidence of profits being overstated in these places as a consequence.

Anyone familiar with the literature on illicit financial flows knows there is no such prima facie evidence: I have seen none at all. I have never heard of people redirecting financial flows and profits to such places. I have never seen a tax agreement that considers the possibility. I have never heard of capital flight of any sort into such places. There is little evidence of profits being reported in developing countries.

I have of course heard of remittances through informal channels: that we know of, but these are not corporate transfer mispricing.

So you have to ask, why should we expect anyone to test a hypothesis, as Fuest seems to require, when there is no evidence that the phenomena exists? And secondly, how can anyone come to a categorical conclusion that such flows are so substantial and significant that they nullify the evidence of massive flows in the opposite direction when no evidence is put forward by them to support that hypothesis? That is what Fuest does. But in that case I have to ask whether his paper qualifies as objective academic observation or as a work of abstract thinking.

Even then Fuest ignores the fact that even if there was transfer pricing abuse into developing countries this would not in any way reduce the abuse that has been found that imposes such substantial cost on them. The transfers in would not match the transfers out. They would (if they occurred, which I doubt) be on different commodities and be by different companies because no one is going to do dual transfer pricing abuse in and out to arrive at a net correct position. So in fact the gains would be entirely unrelated to the losses, could not be netted off and would be a massive transfer pricing loss to developed countries; somthing we contend occurs but which has as its destination the, for multinational corporations, more attractive destination of a tax haven / secrecy jurisdiction. As such the inherent logic of netting off in the Oxford paper is fundamentally flawed whichever hypothesis is right. Remember, tax errors are not calculated net: they are grossed up. And since any errors on inward flows would be to abuse duties these cannot be netted off against outward flows to abuse income taxes.

The reality is in that case that the loss calcualtions stand up to scrutiny whatever happens and may simply be understated. Whatt is clear is that Fuest's critique is wrong and reveals little undertsanding of tax or of the likely nature of illicit flows.

In that case doesn’t it follow that the claim he makes that “tax revenue losses are overestimated drastically” is actually just “wildly exaggerated” or simply “plain wrong“ becasue if there is a risk of error it is in exactly the opposite direction of the one he predict?

Richard Murphy

12:54 pm  

Post a Comment

<< Home