Lots of people will tell you that inheritance tax is unfair. Well, you know what I feel about fairness. I understand the social significance of inheritance taxes, and although it can be hard to shake those misguided feelings of injustice, I'm broadly happy that we have such a tax to prevent cross-generational retrenchment of capital.
No, the problem with IHT isn't that we have it. The problem with IHT is this. If you don't get it, you need to understand that each of the links on that page goes to a subsection with hundreds, if not thousands of pages. You also not to understand that the only reason the document is available is the Freedom of Information Act. And that parts of these pages are redacted under exemptions to that act. And that telephoning the HMRC IHT helpline almost inevitably results in the advice: "tell us everything, we'll tell you what to pay". And that the only real, meaningful alternative is to pay a lawyer hundreds of pounds an hour to advise you.
Like so many other areas of tax law (and law generally), this has become an arena for a cat-and-mouse game played between government specialists and their opposite numbers in the avoidance profession. The only people who really benefit from all this are those who make a profession out of understanding and administering it; the rest of us just have to foot the cost of dealing with all the complexity.
Complexity is not inherently a bad thing. In some arenas, complexity is the necessary corollary of power. Human beings are considerably more complex than single-celled organisms, and consequently have much greater control over their environment. As science and technology have progressed, they too have become more complex; many people find this threatening, given our ever greater reliance on them to provide for our basic needs. But rigorous intellectual practice has a solution to this problem: abstraction. As one builds larger and larger structures, one introduces abstractions that allow one to ignore the inner workings of something when it isn't relevant to the task in hand. Providing that one understands the basic concepts of an abstracted computer interface (currently windows, buttons, folders, files etc), there is no need for most people to get involved with the world of program flow and machine-code instructions, let alone transistors and electrons. One can operate enormously powerful tools without needing to know how they work.
So there is a place for complexity that is well-abstracted and delivers practical benefit. How does our tax system fare on this basis? Well, from where I'm standing, it fails both tests.
There are few effective abstractions in IHT law. Instead, there is a tangled web of individual rules, provisions and exemptions with only loose principles holding the whole thing together. There are few areas that can be understood on their own - each set of rules is heavily interdependent with numerous others, making it hard to gain even a high-level picture without understanding the whole thing.
As for utility... well there's perhaps more of a case to be made there. There is clearly a social function to at least some of the special provisions: agricultural relief, charitable donations etc. But it's also clear that many of the most tortuous changes to the law, particularly as concerns trusts and offshore assets, have been responses to avoidance strategies. The mess is a product of the organic piecemeal growth of our legal system to cope with things that weren't previously envisaged. There is no inherent benefit to much of it; we just can't countenance the upheaval of rewriting it all from scratch.
It is a truism in software development that when a program grows organically in this way, it will eventually need to be refactored: carefully unpicked and put back together again with the introduction of meaningful abstractions to make it comprehensible. If this doesn't happen, the cost of maintaining it becomes prohibitive and it eventually collapses under the weight of its own informational complexity. This process is often unpopular. It tends to remove strange quirks upon which users of the software have come to rely, replacing them with new, more logical constructs. Sometimes it removes large chunks of functionality that are commercially unviable or otherwise undesirable. Reworking any system with a long history and many users inevitably meets with resistance. It's hard to think of many things with a longer history and an impact on more people than our legal system; rewriting chunks of it would anger many people with a vested interest in it continuing to operate as it does now. But at some point it will be unavoidable, because the cost of administering it will be greater than the value it brings. Personally, I think we're probably very close (if not beyond) that line already. And the longer we leave it, the more painful the process will be.