The Invention of Capital

There is amusement to be had in the progress of Google from "Don't be evil" to Doctor Evil's lair, as implied by Eric Schmidt's verbal slip about being a "capitalist country", but this ignores a couple of truths. The first is that the famous motto is widely misunderstood, having nothing to do with ethics and everything to do with the search product's USP of unbiased results. "Evil" here is adrift of morality and no more than a consumption preference. It's just marketing. The second truth is that multinationals got out of state-building with the transition from mercantile to industrial capitalism 200 years ago. That shift was marked by the emerging nation states absorption of the mercantile corporations, most famously with the transformation of the East India Company into the British Empire in India.

The semi-independent nature of multinationals appears to live on in their use of offshore subsidiaries and their ability to negotiate as peers with government over tax, however this is misleading. Tax havens are extensions of the nation state. The British Virgin Islands are no different to the Channel Islands in this respect. Corporations do not operate outside the nation state, they operate inside it. Their exceptional status does not reflect any real difficulty on the part of government in taxing them, it reflects their economic power and thus ability to dictate favourable terms. It's worth bearing this mind as the clamour for tax reform builds. While governments are busy announcing "clampdowns", their policy changes are actually geared to making tax regimes more corporate-friendly.

Though there has been much talk about international cooperation to get tough with tax havens, this is unlikely to produce anything beyond the cosmetic. Tax havens are secrecy jurisdictions, which means they are concerned with personal tax avoidance as much as corporate, and it would take a global revolution to gain agreement to intrude on the financial affairs of the rich. The fact that secrecy jurisdictions have long been the primary conduit for global money-laundering has not led to their being shut down, so why should the embarrassment of Apple or Google change matters? Corporate tax avoidance is relatively small beer compared to personal tax evasion and untaxed criminal proceeds.

The key mechanism behind most high-profile corporate tax avoidance is transfer pricing (aka "fiscal dumping"), which allows a multinational to move costs to higher-tax jurisdictions, and so minimise pre-tax profits, and correspondingly shift revenue to low tax jurisdictions, and so maximise post-tax profits. The original rationale for transfer-pricing was to distribute HQ costs out to operating subsidiaries, thereby recognising the contribution of shared internal services and corporate know-how to local profitability. Over time, both transferred costs and revenues were increasingly made up of intangible assets based on intellectual property (IP), such as software copyrights, patents and the "brand". For example, HQ sells a patent to a low tax subsidiary for a nominal amount. The subsidiary then charges large royalty fees to operating companies in high tax jurisdictions, thus artificially shifting profits to the low tax regime.

There are international accounting rules on transfer-pricing and the valuation of IP, but these operate on an "arms-length" principle, i.e. prices must roughly match what would be charged in a open market by unrelated entities. The problem is that the price of something that is not publicly-traded, such as inhouse-developed software or a patent, is a largely subjective call, much like accounting "goodwill". It is also acceptable to cross-charge for advice, intellectual capital, that again has no fixed open market price. Though this should be done on a transparent cost-plus basis, this simply incentivises the corporation to over-pay executives and specialists and thus ramp up the allowable cost of these services. It also incentivises corporations to pay top-dollar for IP acquisitions (often done through stock rather than cash). A successful purchase can produce large royalties to limit tax exposure. An unsuccessful purchase can be written off against tax. This virtual capital has become a larger component of total capital over time, both as a result of technological change and stock bubbles. Capital is mobile not just in its money form, but increasingly in its constant (i.e. non-labour) form.

So what tax reform can we expect? One proposal is to tax corporate profits where those profits are generated rather than where the business is "tax-resident". If you buy a book on Amazon while sitting at a PC in London, then the profit should be taxed in the UK not Luxembourg or the US. The problem is that this doesn't address transfer-pricing, which can suppress local profits, as was demonstrated last year by Starbucks. The desire to tax the sale will inexorably lead to the tax point inching towards the point of sale, i.e. the actual cash transaction. As Larry Elliott suggested this week, "We can force [government] to introduce sales taxes to avoid profits migrating offshore".

A sales tax is regressive as it falls more heavily on those who have to spend all of their income on current consumption, in contrast to the well-off who typically divert a chunk into savings and investment, as well as non-VATable assets such as property. It is possible to implement it in a way that does not penalise the consumer immediately (the VAT amount is increased and the pre-VAT amount is decreased, with zero net change in price), however the eventual effect is inflationary. In fact, any policy that results in corporates paying their tax in full is inflationary as a subset is probably only profitable today because they currently avoid some tax. Free market theory holds that these businesses are inefficient and should exit the market, making way for new entrants, but in reality some will just put up their prices and get away with it.

Sales taxes are also problematic for government revenues due to widening inequality, which causes median incomes to stagnate, and globalisation, which leads to commodity deflation. Though the latter offsets the former to a degree, it results in a growing proportion of median incomes being spent on non-taxed goods, such as food and rent. This leads to a regular ratcheting-up of VAT and other sales taxes (or their extension to new areas). This is a vicious spiral: widening inequality and globalisation drive increased demand for public services in developed economies while simultaneously depressing tax receipts to pay for them.

Simon Jenkins had much the same thought as Larry Elliott, with a few more thrown in for good measure: "The trouble is that any company whose business is not nailed to British soil seems to treat corporation tax as voluntary. It might be better to ease it out in favour of sales tax, business property tax and, if the City bites the bullet, a financial transactions tax". Business rates tend to be reasonable because they relate to the provision of local services that are proportionate to property, such as roads and utilities. But a general tax based on business property would advantage those with a small footprint relative to profit (such as Google and Amazon), as well as companies that have offshored production. A financial transactions tax is not designed to increase tax receipts for public expenditure but to dampen down risky behaviour (high-volume trading) and to surcharge excess profits that arise (in part) from tax avoidance. In practice, the sales tax would end up doing the bulk of the work in Jenkins's scheme.

One "cunning plan" is to lower corporation tax rates uniformly to a level that stops avoidance being worthwhile - the Laffer Curve for business. The result of this would not be a change of direction, but rather the continuation of the neoliberal strategy of the last 35 years. As Robert Peston noted, the current problems arising are "part of a broad trend of multinationals paying a much smaller proportion of public sector costs in all the world's developed economies. In the US, for example, corporate tax generated 32.1% of all federal taxes in 1952. Today that proportion has fallen to a puny 8.9%". Peston rather ingenuously suggests that "hoarding cash in low-tax centres seems in some ways a bit pointless for publicly owned corporations - in that it creates enormous complications when it comes to getting the cash to its rightful owners, the shareholders". But as Apple has recently shown, these deposits can be used as collateral to raise loans that are then used to pay dividends to shareholders. Thus tax avoidance fuels private sector debt, which is already at high levels due to years of low interest rates encouraging debt financing.

Another proposal is to adopt unitary taxation. This involves rolling up a multinational's profits into a single consolidated number, which is then allocated pro-rata across all tax jurisdictions where it operates based on an agreed formula, e.g. volumes of sales or employee headcount. The ruling tax rate in each state is then applied to that state's share of the profits (it's worth noting that this is already done at a federal level in the US and Switzerland). The biggest benefit is that it makes transfer pricing pointless. However, it doesn't address the pricing of intangible assets. In other words, the consolidated profit figure is still open to manipulation by the over-valuation of IP.

So what's likely to happen? There seems little reason to expect a change in the long-term trend of tax composition in the UK or other advanced economies. This means a continuing shift from income taxes to sales taxes, and simultaneously from taxes levied on specific goods to taxes on general consumption (the standard rate of VAT rose from 8% in 1979 to 20% by 2011). Corporation tax has remained a relatively small component, and highly geared to the business cycle. Local taxes have declined due to the centralisation of government since the 80s. Income taxes have been fairly consistent over time, as a share of total revenue, however this masks a shift of the burden from the well-off to the less well-off through the increasing share of capped NICs, the conversion of income to dividends and capital gains, and the regressive impact of increased personal allowances (i.e. reduced benefit to minimum wage earners). History would therefore point to an increase in VAT as an ironic strategy to tackle corporate tax avoidance.

Unitary taxation has an obvious attraction for the EU, not least because it does not require harmonisation of corporation tax rates, though it would gradually encourage it. Multinationals would see fewer advantages in being based in low tax states like Ireland or Luxembourg, and those states would have less reason to keep rates low as an attraction for multinationals. In practice, there would not be a lot of change. Companies like Google are based in Ireland for a variety of reasons, including language, education levels and local wage rates. There would probably be fewer company secretary jobs in Luxembourg, but I suspect the local economy would cope. The key point that makes unitary taxation a good bet is that it makes no attempt to address the issue of IP valuations. That, ultimately, is what companies like Google, Apple and Amazon are most keen to protect. This would mean that a mechanism remained by which corporations could, in collusion with governments, determine their effective tax rate while burnishing their corporate social responsibility credentials.

It is the gradual transformation of capital from hardware to software that is driving corporate tax avoidance.