Raw Beethoven

Got access to a trial subscription of Amazon's Music Unlimited. Their library has massively expanded since I last looked; seems like most recordings ever made are there now. Like a kid in a candy store. Loads of good stuff, some not so good.
Good. Alfred Brendel playing Liszt, especially the Années de pèrelinage. The apotheosis of Romanticism.
Not so good. Glenn Gould. Playing Bach. I'm thinking, I would literally rather hear this played by a computer.

Of course, computerised renditions of classical music, particularly Bach, have been with us for some time. The original one dates from 1968, Switched-On Bach by Wendy (formerly Walter) Carlos, and now curiously almost unobtainable. But there are good examples on YouTube, just search for "synthesiser Brandenburg".

Via Amazon, I discover there are now computerised renditions of Beethoven sonatas, by an outfit called Berlin Virtual Symphonics working in association with Edgar Höfler. BVS seem to have started life as composers for computer games. I find these robotic versions very appealing; they lay bare the genius of the music without distraction by frills or personal interpretation. Sure, personal interpretation can be nice, and Brendel does it very well, but there is something to be said for stripped-down Beethoven.

The thing works remarkably well (except for trills, which sound like bad data on a CD), presumably because piano sound samples have reached the point of being close-to-perfect. There must have been a lot more to it than merely scanning the scores, and I congratulate BVS and Mr Höfler on an excellent job.

• Long-time readers will know that I take a particular interest in Chopin players. Having got bored with Rubinstein's interpretation of the Nocturnes, I scanned the Amazon catalogue for alternatives. A lot of them are pleasant enough, but the only one that pressed my buttons, in a good way, is by Turkish pianist Fazil Say. Highly individualistic, sometimes a little fast, but faithful to Chopin's Slavonic bittersweet. Mr Say's performances of Mozart are also refreshingly different, and a real joy — though, judging by Amazon reviews, not everyone's cup of tea.

Mr Say is a prize-winning composer, and his moving Istanbul Symphony is worthy of serious attention — a comment I rarely feel tempted to make about post-war orchestral compositions. Incidentally, Say seems to have aroused the wrath of the Turkish government for certain comments deemed to be critical of Islam; you can read about it on Wikipedia.

Brexit: to deal, or not to deal

Should the UK try to have a deal with the EU in place before its departure from it, an event currently scheduled for 29 March? Listening to the warnings of official experts, one would think that not to do so is courting major disaster. Official experts also warned that Brexit per se would be damaging to the UK economy, but so far those warnings seem to have been off the mark.

Whatever deal is struck, and whenever, one thing seems clear. Our politicians must be more than usually careful what they sign up to.

Any gaps or other uncertainties in the wording of an agreement are liable to cause problems down the line for the UK. The European Commission appears to have few scruples about exploiting such uncertainties, and interpreting legal wording 'creatively', if it thinks this would further EU interests. And it is probably not much good expecting support from the European Court of Justice in such a scenario; the Court's track record strongly suggests it will take the Commission's side.

The Commission's case against Ireland/Apple, which has generated a demand for an extra €13 billion of tax from Apple, provides a useful illustration of its approach to matters of law. The fact that there are few meaningful checks on the powers of the Commission means it is able to put its own interests above respect for legal principles such as legitimate expectations.

In the Apple case — currently going through the appeals process — the Commission is applying the law on state aid in a highly innovative way. If this is accepted by the courts, it will significantly change EU law but without the change having had prior approval from member states. Worse, the Commission is applying the change retrospectively, violating one of the key elements of the rule of law.

For details of the case, and its implications, see my article Legal opportunism and the collusion of powers.

Dead Level

Some of the murders are a bit grisly for my taste, but I have otherwise enjoyed Damien Boyd's D.I. Nick Dixon series of crime thrillers. Not least because they contain nuggets of factual information, e.g. the use of "Bronze Commanders" in the forces. I particularly liked Death Sentence which made the Falklands War seem real to me in a way the newspaper stories never did.

Another in the series, Dead Level, is interesting because it goes into the human-vs-animal insulin question in some detail. It seems that the insulin which sufferers of Type I diabetes have to inject to stay alive is normally "human insulin" — though the "human" reference is misleading since the stuff is manufactured in the lab using microorganisms. A significant proportion of diabetics seem to get unpleasant side effects from this type of insulin, and do better with the animal variety which is extracted from beef or pork. However, it appears that many are not made aware of the possibility of a choice, being given the absolute minimum of information, and as a result some suffer needlessly.

The InDependent Diabetes Trust, which makes a fictionalised appearance in the novel (as the good guys), has an interesting article about this. The article advises not abdicating your own judgment by simply assuming that experts must be right, especially when those experts are driven by considerations other than your preferences. Dead Level follows the standard convention of fiction that when doctors turn bad it is due to financial motives, but there are other reasons why medical professionals can be dangerous to your health.

The diabetic community cannot survive without this life-saving drug. We therefore form a captive market and, as any economist will tell you, this creates a perfect opportunity for experts to manipulate and exploit us. Have our gurus the time or inclination to guide us through this jungle of short, medium, long-term and mixed insulins? Do they explain the scientific jargon and help us make a free and informed choice? [...]

Three clues will help us make up our mind. The first is not to abdicate in favour of the care team or encourage them to steal our melody. The second clue is not to make a decision until you have in front of you a complete list of all insulins on the market, both animal and 'human'. Thus fortified, you and your helpers can work out what is possible for you. An informed, rather than an imposed, choice will result. The third clue is not to believe what the printed instructions tell you about the strength and duration of action of each kind of insulin.

I prefer to use the term "preferences" in relation to patients. As I pointed out in The Power of Life or Death, the phrase "best interests" is often misused, for example when doctors or judges claim it is in a patient's best interests to die — even when the patient didn't express a wish to do so. The modern state-remunerated doctor may well be incentivised to keep his patients in the dark, and may indeed feel morally justified in doing so.

Credits where credits are due

It's that time of the year again for putting one's hand in one's pocket, to make a contribution to a deserving cause.

I refer of course to Wikipedia.

Eight years ago I suggested that Wikipedia would emerge as the most useful product of the internet, and I stand by my forecast. (Facebook? Twitter? Guardian's Comment Is Free? Ahem.)

Of course Wikipedia isn't perfect; it has some flaws, and sometimes needs to be taken with a pinch of salt. It suffers from occasional biases, depending on the topic — though it doesn't seem to me to have an overall bias in any particular direction.

Its open-editing model has worked extraordinarily well, something that would have been hard to predict twenty years ago. (I wouldn't have guessed it could, before seeing it in operation.) Founder Jimmy Wales deserves enormous credit for having had the vision to see the model through to fruition.

If Wikipedia had ads it would be worth billions, but it remains steadfastly non-profit.

Here is a link for making a donation to parent body Wikimedia:
https://payments.wikimedia.org/index.php?title=Special:PaypalExpressGateway&appeal=JimmyQuote&ffname=paypal_ec&recurring=¤cy=USD&amount=0&payment_method=paypal&uselang=en

Honourable mention: Internet Archive, which contains a ton of useful stuff and, like Wikipedia, provides its services to the public for free. At the time of posting this, an anonymous benefactor is matching donations to Internet Archive, dollar for dollar.

Apple and the "Double Irish"

One source of comparable data for Apple seems to have been ignored by the European Commission in their Decision against Ireland/Apple, unless they considered it best not to mention it. Apple's tax scheme is remarkably similar to the "Double Irish" arrangement, a tax minimisation device apparently used by a number of US multinationals, including (allegedly) Facebook, Google, IBM and Microsoft.

It seems reasonable to suppose, based on the information publicly available, that it was Irish tax policy to permit the Double Irish for any multinational group that sought to avail itself of it. This points to the possibility that the Apple rulings were not anomalous, but reflected standard practice. It is odd that this source of data for settling the question of whether there was selective treatment, and hence "state aid", is not mentioned in the EC Decision. It is difficult to imagine that Commission investigators were not aware of it.

Apple's corporate structure in Ireland appears to have differed in one respect from the standard Double Irish structure. Instead of using a two-company structure, Apple used a single company which was deemed split into (a) tax-resident branch and (b) non-resident head office. (It used this structure for each of two companies: Apple Sales International and Apple Operations Europe.)

The EC Decision spends some time arguing — not in relation to the Double Irish question but a different issue — that a branch has to be treated differently than a company. It is doubtful whether this is correct in a transfer pricing context, where one generally has to treat a branch as if it were a distinct legal entity. However, even if using one company rather than two affects the position regarding state aid, to omit any mention of the Double Irish in the Decision seems negligent.

Edited extract from 'legal opportunism and the collusion of powers'

Germany and the EC

There is a popular theory that the EU is run by Germany and/or France.

Possibly the European Commission is dominated by a coalition of individuals having disproportionately German or French nationality. However, the idea that the EU state is controlled by a subgroup of nations seems outdated.

In 2017, Germany's Ministry of Finance published a highly critical review of the EC's approach to state aid. The publication pleads for restraint in applying the state aid rules to taxation, and talks about a need to respect legitimate expectations.

At one point the authors refer to a "mutual understanding among member states [not to] distort trade". The message seems to be: state aid legislation was just a means for us to agree to respect the ideal of the single market. We made it up, so we shouldn't have to suffer from it — at points where it doesn't match our objectives.

The notion that the EU is still just a civilised club of countries, which can agree to restrain the state aid law juggernaut, seems appealing but overly sanguine.

The EU apparatus has moved well past the stage of acting simply as supervisory authority. It is now a fully fledged state in its own right, with its own agenda, and the power to order individual countries to change their behaviour. It is probably beyond control by any subgroup of nations, let alone any one nation.

Edited extract from 'legal opportunism and the collusion of powers'

The weak case against Apple/Ireland

The final part of EC v Apple is on the website:

‘legal opportunism and the collusion of powers’

Tax rulings and the EU

The fact that transfer pricing is more of an art than a science generates a good deal of uncertainty for multinational corporate groups with regard to their future tax liabilities. A group may decide on a transfer pricing methodology which it considers reasonable, and which its tax advisers agree is acceptable. But the risk remains that a national tax authority will disallow the methodology, possibly years after the event.

One way for a group to reduce uncertainty is to try to obtain a "tax ruling". This is a written clarification, issued by a tax authority to a taxpayer, about how tax law will be interpreted in a particular situation. Such rulings often last for a number of years, protecting taxpayers from the risk that an authority will unexpectedly change its interpretation.

Tax rulings are used in areas such as transfer pricing where there is uncertainty about how authorities will apply taxation provisions. They facilitate businesses' efforts to engage in long-term planning, and so tend to promote economic efficiency.

According to the European Commission, tax rulings could be used to provide covert subsidies to businesses, by providing abnormally favourable interpretations of tax law. However, even if in principle tax rulings fall under the remit of state aid law, there are obvious reasons why the Commission should stay clear of this area. State aid legislation is already at odds with the rule of law:
• it violates legitimate expectations by allowing one government authority to overrule the prior commitments of another government authority, to the detriment of individuals; and
• it requires the overruling to have retrospective effect.

To extend application of this to areas where there is already uncertainty about interpretation of the law is to compound existing wrongs. Whenever a tax ruling is issued, there would be the possibility that the Commission subsequently argues in favour of a different interpretation of the law, and demands retrospective adjustment of tax liabilities. Apart from violating the rule of law, this destroys the purpose of rulings which is to reduce uncertainty.

'EC v Apple: collusion of powers' will be published on the website on 29 October.

Legal opportunism

"Creative accounting" is the exploitation of loopholes in accountancy rules to achieve objectives that deviate significantly from acceptable practice.

There is an analogous type of behaviour in matters of law, sometimes called legal opportunism. This involves exploiting areas where the law is vague or ambiguous, or where it has unintentionally left gaps. Imperfections in the law are used to facilitate actions that were not intended to be sanctioned.

It is not only private individuals and corporations who can indulge in legal opportunism. Government may in theory do so, by applying the law in unexpected ways. For example, a national government may occasionally resuscitate an ancient statute not used for centuries, or apply a loosely worded law in an innovative way to target some unenvisaged menace.

By and large, however, Western governments do not engage in this kind of action. One reason is that, since they have the power to create statute, they do not need to resort to opportunism. There is always the risk that a senior court may rule an opportunistic application of a law to be invalid. A new law does not carry that risk.

Another reason is that Western states generally try to maintain a reputation for neutrality vis-à-vis the law. A government with a cavalier attitude to the law as it stands tends to be one that is not held in high regard by world opinion.

For purposes of illustration, take the case of a corporate merger being considered by a national competition authority. The authority typically invites interested parties to submit their comments on the case. In reaching its conclusions, the authority is likely to make an effort to interpret prevailing laws and principles impartially — or at least to be seen to do so.

On the other hand, parties who submit opinions about the merger, whether for or against, are not expected to be neutral. Their arguments may involve mainstream economic analysis, but if the approach is slanted in the interests of one side or the other, no one is going to be very surprised, or necessarily think less of them.

By contrast, a government body that gets openly creative or partisan in its interpretation of the law, in order to further its own agenda, loses its claim to neutrality and ceases to be trustworthy.

The above is an extract from the forthcoming article 'EC v Apple' (part 3), scheduled for publication in October.

Do courts act as a restraint on government excess?

There is always a risk with government that it will attempt to exceed its legitimate powers. However, in jurisdictions where the state structure includes an independent judiciary, such attempts can in theory be held in check by the courts.

The notion that the state should be modelled so as to incorporate sources of internal opposition — often referred to as "separation of powers" — is a relatively modern idea. It depends on taking a sophisticated view about what government should be: not just a device for achieving what is desirable, but one with built-in restraints which on the face of it make its job more difficult.

The courts acting as a check on government requires the presence of a legal system that is autonomous enough, and self-confident enough, to be capable of critically comparing what the government tries to do with what the rule of law demands.

Notional autonomy may be a necessary condition for the role of constraint on government but is probably not a sufficient one. Opposing the government means regarding something other than government as your master; adherence to abstract legal principles, for example. If your primary commitment is to an agenda which happens to be the same as that of the government, you are likely to have little motive for resisting the government when it tries to bend the rules.

The legal system of the EU, including the European Court of Justice, was created from scratch along with the corresponding governmental entity (the forerunner of the European Commission). The whole structure was aimed at achieving a particular objective, namely European integration. A system of this kind is likely to be compromised from the outset with regard to judicial independence. The unspoken purpose of the legal apparatus will be to assist in moving towards the objective, not merely to enforce the prevailing set of rules.

Not surprisingly therefore, there is a question mark over whether the European Court of Justice is genuinely independent, or adequately fulfils the separation-of-powers function.

The above is an extract from the forthcoming article 'EC v Apple' (part 3), scheduled for publication in October.

Transfer Pricing for beginners

Transfer pricing is an issue that arises for multinationals, i.e. corporate groups with a presence in more than one tax jurisdiction. Total group profits have to be allocated among the individual companies, and this is achieved by the group setting notional prices for goods or services transferred from one group company to another.

Say a company makes motor cars in country A and then sells them via its subsidiary in country B. The profit which the B‑company reports for its selling operation will depend on the amount charged by the group per car transferred to it. The lower this transfer price, the more of the overall profit will be allocated to the B‑company. (See figure below for an example.)

Some groups may try to bias profit allocation towards countries with a lower tax rate by adjusting the transfer price. If country B has a higher tax rate than A, it is in the group's interests to minimise profits in B by setting a relatively high transfer price per car.

Governments are of course aware of this ruse. B's tax authority is likely to reject the company's calculation of profit if it considers the transfer price unreasonably high. Conversely, country A may object if it considers the price to be artificially low, so that too little of the profit is taxed in A. Disagreement about whether transfer prices are reasonable can result in long-drawn-out disputes.

The principle which guides authorities' decisions about whether a transfer price is reasonable is the arm's length rule. This states that transfer prices should approximate to what would happen in transactions between unconnected parties. Unfortunately, the interpretation of this is often anything but straightforward. There may be little or no market data to help answer the hypothetical question: what transfer price would be set if the two divisions were independent?

In the case of relatively homogeneous industries like motor cars it may be possible to look at mark-ups on cost charged by rival manufacturers to independent distributors, take an average, and adjust for any special factors. Alternatively one can look at the profit margins of independent distributors, and argue that the transfer price should result in a similar level of profit margin for distribution companies within a group. These are some rough and ready methods used in simple cases.



At the other end of the spectrum, if what is being transferred within the group is the use of intellectual property ("IP") — say from a technology company to one of its subsidiaries — there may be little or no basis for calculating an arm's length price.

A method often used in market-based transactions involving IP is to charge the IP user a percentage of the revenue he obtains from selling products that make use of the IP; in other words, a royalty. This suggests that the IP holder within a group should charge a royalty to other group companies. But royalty rates used in market transactions range from less than 1 percent to more than 20 percent, so this does not necessarily get one very far.

The above is an extract from the forthcoming article 'EC v Apple (part 3)' which is scheduled for publication in October.

LEGO and sex

I've been monitoring the LEGO cards that come free with Sainsbury deliveries. They're funny! But there's something curious about the gender division of the characters.



There seems to be some kind of pattern here. I haven't quite figured it out — but I'm working on it!

coming soon

 
● New article is in progress.

Interesting times

An interesting phenomenon of recent times is how a significant number of politicians, ostensibly committed to the notion of democracy, demonstrate their commitment to another set of values that overrides their support for majority voting.

Thus we witness, for example, a White House member of staff boasting anonymously in the New York Times* about his (or her) efforts to subvert US President Trump's policies.

The anonymous writer is vague about what is so objectionable about the policies that it justifies such behaviour, though he tries to paint a picture of a President who is erratic and deluded. The writer summarises his complaint in terms of Trump's alleged "amorality", by which he apparently means that the President "is not moored to any discernible first principles that guide his decision making". I can see why someone might dislike lack of commitment to principles in a politician, but it is hard to understand why it should be grounds for subversion.

A Republican commentator has expressed disapproval** of the article, arguing that the writer's behaviour will only make the President "more defiant, more reckless, more anti-constitutional, and more dangerous". The commentator recommends that the President's aides should rather resign, or work towards impeachment.

The commentator might have suggested that the writer, and like-minded colleagues, should attempt to overcome their antipathy and comply with the President's wishes as far as possible, on the basis that Americans voted in a way that implies they would like them so to comply. However, the commentator does not do so.

This revealed weighting among elected politicians and commentators between (a) the results of a vote, and (b) their own personal preferences is surely one of the most significant contemporary issues in political affairs, and worthy of attention.

Exercise for the academically inclined
1. See how many recent scholarly articles you can find on the phenomenon of Western politicians opposing the results of majority voting, in which the authors regard it as something that needs explaining.
2. Contrast the number you get in part 1 (which may well be zero) with the number of articles which ask how it came about that voters recently made 'bad' decisions and which suggest remedies for preventing it from happening again.

* 'I am part of the resistance inside the Trump administration', anonymous op-ed, New York Times, 5 September 2018
** David Frum, 'This is a constitutional crisis', The Atlantic, 5 September 2018

---

● The third and final part of EC v Apple should be on the website later this month.