Tax cases are expensive and complex to investigate. Google probably has more money that the entire UK tax authorities. HMRC has to concentrate their efforts somewhere, and they concentrate on the people who are clearly breaking the law, or where they have whistle-blower evidence that laws are being broken.
That puts companies like Google at an advantage. They can operate at the edge of the law, exploiting loopholes, and then pay more if the laws are changed. And then, if we find out that they are not obeying the law but breaking it they make a deal with HMRC. See the deal Vodafone made where a huge tax bill was waived, if Vodafone paid a small part of the tax they had evaded. (http://www.guardian.co.uk/money/2011/dec/06/hmrc-tax-deal-vo...)
Dodging tax is only beneficial to these companies.
As to why they should pay: Apart from the fact that UK customers of Google buy UK ads for their UK potential customers using UK bank accounts and UK pounds and buy them from UK staff at Google's UK office (with a final 'close the sale' bit of paperwork done in Ireland for the sole purpose of avoiding tax in UK) - apart from that, Google makes use of UK schools and roads and police and fire departments and etc.
No one minds that Google pays less tax than they should. People only mind when Google goes to extraordinary lengths to avoid tax.
I agree with paying the minimum amount of tax but what does that actually mean? Eric Schmidt said he believes taxes should be at the base of economic activity (ie Ireland) but this is bullshit because it puts all the power in the hands of the corporations and incentivies countries to lower taxes to encourage corporations to move there.
Taxes should be based on where products are consumed so that each country taxes the exact percentage of profits generated from sales in that country. If in the UK Google sales amounted to £1b profit then £230m should be corporation taxes.
Secondly, it's disingenuous of them to ask the taxpayer to "just change the rules" if we don't like their avoidance, when they're funding the very regulatory capture which makes it hard for us to do that, via their investment in these schemes.
don't be evil is not the same as do no evil
Expect the law to change, creating an even more complicated tax structure and destroying even more economic value while providing zero benefit to common citizens.
We should be looking in the opposite direction. Do away with large swaths of the tax code and ask less of the companies that support the global economy. Especially now that apparently the organizations that extract those taxes (specifically the IRS) have become political tools to harass those people whose politics our "leaders" disagree with.
I'm in favor of the tax-code placing an equal burden on everyone. I am against only corporations getting a free-pass though.
If government didn't have the scale to take on such projects a lot of these corporations wouldn't exist and a lack of research like those often government funded would just reinforce the status quo.
The fact is, that if government hadn't been employing these engineers to create war technologies, our technology would probably be much more advanced and the world would have less war and be a much better place.
Whenever government employs people to do something, it necessarily keeps them from doing something else. This is well explained in chapter 4 of Henry Hazlitt's Economics in One Lesson (http://steshaw.org/economics-in-one-lesson/chap04p1.html) in taking the example of government spending for a bridge:
"The bridge exists. It is, let us suppose, a beautiful and not an ugly bridge. It has come into being through the magic of government spending. Where would it have been if the obstructionists and the reactionaries had had their way? There would have been no bridge. The country would have been just that much poorer. Here again the government spenders have the better of the argument with all those who cannot see beyond the immediate range of their physical eyes. They can see the bridge. But if they have taught themselves to look for indirect as well as direct consequences they can once more see in the eye of imagination the possibilities that have never been allowed to come into existence. They can see the unbuilt homes, the unmade cars and washing machines, the unmade dresses and coats, perhaps the ungrown and unsold foodstuffs. To see these uncreated things requires a kind of imagination that not many people have. We can think of these nonexistent objects once, perhaps, but we cannot keep them before our minds as we can the bridge that we pass every working day. What has happened is merely that one thing has been created instead of others.
The fact is that a businesses number one aim is to make a profit and that restricts most businesses to seeing no further than the end of their nose and limits their capacity for risky investments. When it comes to research, corporations should be looking for profitable applications of research before anything else because it's their number one aim. This means that a lot of research that doesn't have any clear practical application to a profitable model or doesn't meet a percieved profitable timescale is unlikely to be pursued.
I'm not denying that some companies do some groundbreaking research just that the inherint nature of corporations prevents them from taking risks that governments can take and which a lot of corporations capitalise on. Corporation and government funded research are two sides of the same coin. They both bring their respective benefits to society but one won't work without the other.
Most corporations would never have taken the big initial risks and government would likely never have practically applied the findings to a valuable product.
For example, a government should not build a road unless the people using the road will collectively save more in transportation costs than the costs to build the road.
The problem is that for many projects, the net benefit is unknown at the start. In the road example, before a road is built, it may not be clear how much it will cost to build a road nor how much people will save from using the road. The best we can do is try to estimate.
But just because we have to estimate doesn't mean that we ignore profit. Before a project is undertaken, we should still try to figure out if its worth doing. We just can't be sure.
When we say that government funds "risky" projects, this can mean that the projects have a high probability of failure but are still worth funding because the probabilistic expected payout is positive, or that the projects are really not worth funding because we really expect that there will be no net benefit. If we're talking about the first, then the private sector will fund it without government. For example, YCombinator routinely funds companies that have a high probability of failure. But overall it expects a positive payout. So startups get funded.
But if we're talking talking about the second type of risky, as in no expected return, then really no one should be funding these projects. Why should the taxpayers take risks collectively that no one is willing to take individually?
It doesn't make sense for an individual to try to fund retirement by buying lottery tickets, no matter how much money they spend. Therefore, government should not be trying to fund government pensions by buying lottery tickets.
In practice, governments seldom openly advocate investing in projects with negative returns like a lottery. They propose funding projects that the private sector won't fund. But even if a project is expected to be profitable, it still should not be funded at the expense of projects with higher expected profit. And because we have limited resources, the funding of low-expected-profit projects means that other higher-profit projects can't be funded.
Again, I refer you to Henry Hazlitt's Economics in One Lesson, chapter 6 this time: http://steshaw.org/economics-in-one-lesson/chap06p1.html.