This weekend's G20 Summit declaration included a worrying paragraph on tax. The part about which I am concerned (which opened the paragraph) said:

We will continue our work for a globally fair, sustainable, and modern international tax system based, in particular on tax treaties and transfer pricing rules, and welcome international cooperation to advance pro-growth tax policies. Worldwide implementation of the OECD/G20 Base Erosion and Profit Shifting package remains essential.

Why is this worrying? There are two aspects of concern.

One is the blind commitment to growth through trade. This is in such conflict with COP24 that it is deeply concerning.

The other is the commitment to the Organisation for Economic Cooperation and Development's Base Erosion and Profits Shifting process. Let me share this on BEPS, which I have co-authored for submission to the IMF, very soon:

The BEPS process was not without its merits, the biggest of which we would consider to be the recommendation of country-by-country reporting. We would simultaneously suggest that the OECD's promotion of systematic automatic information exchange between countries is of great importance. However, we think there were also major weaknesses in the process. In particular, it assumed that the existing structure of international tax relations and the basis on which multinational corporations should be taxed should remain intact. This means that:

• The weaknesses that exist in the structure of international tax agreements continue to exist;
• More importantly, the OECD's arm's length basis for the taxation of multinational companies has remained intact. This means that these entities continue to be taxed on the basis of the fiction that they are made up of independent entities, the existence of each of which should be assumed to be commercially justified, when there is significant evidence that this is not true.

Our suggestion is, then, that what OECD BEPS did was place a ‘sticking plaster' on the existing system of international taxation rather than stand back and ask what the real weaknesses and spillovers within the existing tax system were, which would have identified the existing tax treaties and the arm's length pricing method as issues to be addressed, rather than to be retained. We feel, then, that there was a systemic failing in the BEPS process.

We elaborated this later in the submission as follows:

The current international system for taxing corporations, in particular, is based on an economic and legal fiction. This fiction is that all the entities that make up a multinational group of companies should be considered to be independent, commercially justifiable corporate entities whose transactions are all entered into for the sole purpose of pursuing its own profit maximisation, which task it fulfils by trading at market prices for the benefit of its shareholders, to whose identity it is, however indifferent.

This assumption is not true. The subsidiary companies are not indifferent to the identity of their shareholders: they exist solely to serve those shareholders, purpose, which need not be the maximisation of the profit of the entity in isolation. In addition, they act solely in accordance with the instructions provided by those shareholders. If that means that the company does not maximise its profit, or does not trade at market prices, but for the benefit of the group as a whole, then that is what they will do. That is the actual purpose for which they exist. And for that reason there is no requirement that a subsidiary has any commercial substance at all, at least in its own right: it may exist solely to ring-fence a liability, for example; or to delineate an activity for purely regulatory purposes; and it might just as easily exist solely to assist the mitigation of a tax liability for the group of which it is a member.

The presumption that arm's length pricing is in any way appropriate in this circumstance, or that there need be a separate profit motive, is wholly inappropriate. If that profit motive exists it is driven by the board of the corporate entity as a whole, which will not, by any means, record all the income arising in the entity on the basis of separate entity accounting.

The accounting profession and company law did recognise this fact a long time ago: consolidated accounting for corporate groups has been commonplace for about 70 years. Nonetheless, despite a steady move towards basing taxation liabilities on accounting profits this fact has not been reflected in tax law.

This has largely been the result of taxation politics. There has been a failure at an international level to agree to a change to the system of taxing rights over multinational companies created by the League of Nations in the 1920s and 1930s. It is important to note that at this time consolidated accounting was rare: the data to tax on a group basis did not always exist. It was unsurprising that a separate entity method of taxation was adopted in that case. But since the practice of accounting has moved on many decades ago it no longer makes sense, barring the fact that the national politics of this issue, and the practices of the OECD have been heavily influenced by extensive lobbying from those with a vested interest, to maintain the status quo.

Those vested interests include:

• multinational corporations themselves, who have clearly benefited from the ability to avoid tax that the system has created;
• the tax havens whose well being is at least in part based on the relocation of corporate profits to these locations, with their credibility and legitimacy being even more closely associated with this activity; and
• the firms of accountants (in particular) who have made it a specialist business to advise on the creation of so-called ‘transfer prices' that supposedly reflect market prices when there is, in fact, no market in existence for many of the activities that take place for the internal trades recorded by the subsidiaries of multinational corporations. The profits that they make from being the controllers of the vast majority of the intellectual property associated with this activity are likely to be prejudiced by any change to this tax basis.

It should be noted that similar problems exist with regard to the source and residence basis of taxation when it comes to corporations. Whilst source bases of taxation can be hard to dispute in some cases (for example, the extractive industries) tax residence is largely a matter of choice for the subsidiaries of multinational corporations, and even those corporations themselves. This opportunity is extensively gamed at cost to all nations, but most especially source states, against whose interests the standard OECD double tax agreement is biased, particularly with regard to limitations in tax withholding. This gaming undermines the credibility of the system, whatever the intention behind its creation. Further tax spillover effects are created.

The means to tackle these spillovers exist. Country-by-country reporting has indicated the possibility of this. The fact that it (uniquely) originated in civil society and has not in any way been endorsed by the accounting standard setting establishment and has been resisted by the largest firms of accountants and auditors does indicate that the obstacle to progress that tax professionals present as noted above is real. Country-by-country reporting (CBCR) does, using the minimum number of necessary variables to indicate economic activity (which might, it should be noted, need to be extended in the case of extractive industries activity and maybe banking), suggest where it is likely that the economic substance of activities is located. Ideally CBCR would report by country:

• Sales by country, separated into both third party and intra-group transactions, on both a source and destination basis;
• Labour, both by headcount and total employment cost including the cost of benefits in kind and secondary forms of payment such as share options;
• The cost tangible asset investment by location excluding intra-group balances;
• Shareholder funds;
• Profit before tax;
• Current tax due;
• Current tax paid.

These elements can then be used to apportion the profits of a multinational corporation to states by formulaic calculation. The question of profit apportionment is then resolved. The consequence is that the state to which the profit is then allocated is liberated to charge whatever tax rate it wishes. This, it should be noted, would not end tax competition. States might still offer low corporation tax rates to induce the inward relocation of labour or tangible investment, in particular, but at least tax competition might then take place on the basis of the economic substance of transactions, as both market practice and economic theory might suggest appropriate. This is the optimal solution for resolving the problem of apportioning taxation rights.

Alternatively, and as an interim step, states might wish to adopt an ‘alternative minimum corporation tax'. This would require a calculation based on available CBCR data, as noted above, to determine the proportion of total multinational group profits (declared on an accounting basis if not otherwise capable of determination) attributable to a jurisdiction. This would then be subject to a deemed effective tax rate that might be a high proportion of the standard corporation tax rate of the jurisdiction in question. If the resulting sum was higher than the tax due based on declared profits then it is suggested that the difference be charged as an excess charge. It is likely that this would only be an interim step whilst a full apportionment basis for corporation tax was agreed internationally to tackle the spillover effects the current system creates.

Despite an awareness of these issues - and the IMF would not be asking the questions on BEPS to which I have responded unless they were aware of the criticisms of BEPS -the G20 remains committed to the taxation of multinational corporations on the basis of a fiction. And that makes no sense at all.