Free Education? Don’t make the situation worse!

We love to moan about the system – how it conditions our thought, places expectations upon us, is inflexible and ill-suited to the modern context etc. – and that moaning isn’t limited purely to students. Free education sounds wonderful but, in reality, a subsidised higher education sector works against students’ best interests.

The increasing supply of universities, places, graduates, qualifications etc. continuously devalue educational qualifications. With the exception of courses that have a significant vocational content such as Medicine, Engineering, Nursing, Teaching and Natural Sciences, many graduates will find their course’s academic content mostly unnecessary for the line of work they plan to enter. Unfortunately, an oversupply of graduates means that many firms advertise relatively well-compensated occupations as being exclusively ‘grad jobs’. This serves to reinforce the perception that you actually need a degree to even be capable of doing these jobs when, in actuality, it’s just that so many people currently have degrees that it’s pointless applying if you don’t. The necessary skills are better taught outside of a university.

What about all those who would essentially be coerced into going to university because, with free education and the increased supply of graduates, they’d have even less of a chance out there without a degree than they do now? What about those who left education earlier and whose relatively meagre qualifications are further devalued because of more graduates in the labour market? Funnily enough, the income inequality that education subsidies purport to alleviate would only increase. The training required to get a ‘good job’ (and, therefore, to fill them) would simply be lengthened due to qualifications’ devaluation. Normative signposting for how best to spend time is a subtle deprivation of civil liberty.

What is education? Why do we value one form of learning over another? Why stop at higher education? Why not subsidise gap years to Southeast Asia where people ‘discover themselves’? Subsidising one form of education almost always forcefully elevates it to a normatively superior perceived status; this perpetuates social structures, labour market characteristics, outcomes etc. since this normative dimension of legal institutions works to resist our attempts to reinvent social structures and deviating from the accepted norm. Does society really need to pay to offer free behavioural conditioning and thereby limit its own evolution? Free education protests are (mostly) unintended expressions of backward, socially destructive and misery-perpetuating conservatism veiled in social liberalism via equal opportunities and rights rhetoric.

Is private currency history repeating itself?

As I and others have said before, free banking & competitive currency provision either completely ignored, or unfairly dismissed by opponents who really don’t know too much about the system. A recent example came on the FT’s Alphaville blog.

By author Izabella Kaminska, and entitled “Private money vs totally-public money, plus some history”, it purported to show how cryptocurrency was just a rerun of earlier monetary struggles, looking specifically at the formation of the Bank of England:

As the BoE’s historical timeline helpfully points out, the BoE came into being when a private syndicate decided to risk all in 1688 by providing the UK government with funding when no-one else was prepared to do so. This ultimately proved to be a very good decision. It turns out lending money to government on terms you can enforce and control can be very profitable, especially if it leads to wise public investments that improve the wealth of the nation and make it easier to collect taxes as a result.

Soon enough the Bank’s success meant it could raise financing for both the government and private interests from almost anyone, issuing notes and deposits to all those who were prepared to do so.

Before the Bank knew it, its notes had become the most liquid and trusted in the land.

Open and shut then—free banking evolves naturally into superior central banking! Or maybe not. As George Selgin pointed out in the comments.

Ms. Kaminska goes out of her way to dismiss the famously efficient and stable Scottish system as an “oligopoly” without even bothering to offer any evidence that the banks in that system colluded or otherwise behaved differently than they might have had entry into the industry been unrestricted. (For her information on the Scottish system Ms. Kaminska relies on a single blog post that in turn draws on some untrustworthy sources, happily ignoring the extensive literature on the other side of the question.)

Ms. Kaminska then imagines that the English system’s only fault lay, not in the dangerous concentration of privileges in the Bank of England, awarded it not owing to any enlightened concern about stability but simply in return for its fiscal support of the English government, but to the fact that that monopoly was as yet not complete!  In fact a currency monopoly is extremely dangerous because it immunizes the monopoly bank from the normal discipline of routine settlement, making it capable of acting as a sort of Pied Piper to less privileged banks.  (Peel’s Act itself, in turn, caused trouble by undermining the English system’s ability to accommodate changes in the British public’s demand for currency.)

In light of these facts, Ms. Kaminska’s claim that English banking crises were caused by smaller joint-stock note issuing rivals, which were at last allowed to compete with it, albeit only outside of the main, metropolitan market, beginning in 1833, is utterly–I was going to write “laughably” except there’s nothing fun about it–mistaken, as she might have discovered had she bothered to read, say, Walter Bagehot’s Lombard Street, say, instead of copying and pasting from the Bank of England’s own self-serving web pages. She would there have come across Bagehot’s careful account of how England’s artificially centralized “one reserve” system, dominated by the Bank of England in consequences of its accumulated privileges, exposed it to financial crises to which Scotland and other less centralized (‘”natural”) banking systems were immune.

The winds of political change

UK by-elections (like last week’s in Rochester and Strood, where the UK Independence Party gained its second MP) have always been an opportunity for electors to vent their contempt for the national politicians, before things return to normality at the general election. By-elections generally do not matter; general elections do. So voters’ actions are perfectly rational.

But few people, even the pollsters, are predicting that things will return to normal at the general election in May 2015. Though Scotland did not vote ‘Yes’ to independence in its recent referendum campaign, the performance of Labour, the main ‘No’ campaigners, was humiliatingly poor. But the Scottish National Party is now piling on support. It now has 90,000 members – roughly half the number that the Conservative and Labour parties are able to achieve, even though their UK-wide base is twelve times larger than Scotland alone. Again, the SNP has often done well in by-elections, but never managed to break through in UK national elections. But now there is a real feeling that normality will not return this time, and that the SNP will steal anything up to 40 Westminster seats from Labour.

The Liberal Democrats, the Conservatives’ coalition partners in government, are meanwhile being humiliated just about everywhere. In the Rochester and Strood by-election, they lost their deposit for the eleventh time running, polling just a few hundred votes. Their core supporters think they have sold out to the Conservatives, while voters who want to send a rude message to Westminster have thought UKIP a much better way to do that. In the past they voted for the LibDems, but now the LibDems are part of the Westminster establishment that they despise.

The main parties, then, find themselves no longer leading the agenda; what will decide the election is how these minor parties fare in May 2015. But this phenomenon is not unique to Britain. All over Europe, minority parties are shaking the political class and winning footholds in the legislature.

What is going on, and why? Perhaps we have to look outside the political process to understand. In commerce, for example, traditional business models have been fundamentally disrupted by the internet. Retailing in particular has been rocked by new suppliers, new ways of shopping and new delivery systems. With things like Amazon Click & Collect, why do we need a Royal Mail – even a private one, as it is now. And much the same is happening in politics too. Small communities can find each other, and organise and mobilise, and cause real problems for the traditional parties.

Given today’s technology, there is no reason for people to settle for off-the-peg goods and services. They can be made to your specification, and shipped direct to your door. Barriers to entry have been swept away, as new suppliers with new ideas and not much more than a website can suddenly enter the market and challenge the incumbents.

It is the same in politics. When people have a choice of umpteen different TV or phone or utility packages, they become increasingly contemptuous of national and local government ‘take it or leave it’ services. When Air B&B or Über enables people to access services in an instant, they wonder why they have to fill in forms and queue up in council offices. What is the point of a Met Office when you can get the weather on your phone from countless other providers?

And national parties find it harder to dominate the national debate, as newspaper sales have been falling, because more and more people get their news from online channels – and not necessarily from the traditional media companies, but from a huge number of new media channels, plus (increasingly) social media and other sources. Activist groups can find each other and mobilise. The domination of traditional media and traditional parties is being eroded by people power.

Through internet and communications technology, we can also bypass government services more easily. Telephones were a nationalised industry thirty years ago, but nobody even thinks about re-nationalising them today. And given the new multiplicity of information and entertainment channels, more and more people are asking why we really need the BBC – that one-time flagship of the British establishment – as a state broadcaster.

The internet also makes it easier to find a private doctor or a private tutor, or indeed to find a job and an apartment. Self-help groups provide help to patients or parents that the lumbering government systems simply cannot provide. Who needs government?

Not many of us, any more. Nearly as many people in the UK (176,632) told the census that their religion was Jedi than there are currently members of the Conservative Party. With falling memberships, party candidates are becoming increasingly irrelevant to most people. They are chosen by a dwindling core of of grey-haired Conservative activists or hard-line-socialist Labour ones, with outdated, intolerant or patronising policies to match.

The politicians’ response has not been to understand these new trends (their attempted use of social media is, as we have seen recently, usually disastrous) but to insulate themselves. Politics is no longer something that successful people in other fields did for a few years as a service to their country, but a full-time career, carefully preserved as such.

No wonder people are upsetting their applecart. And no wonder that they cannot understand why.

Are our children safe with Children’s Services?

The great majority of those who work in local authority children’s services do the very best they can in very difficult conditions.  They suffer abuse from the parents with whom they deal and the media, poor management, criticism from Ofsted and, from their perspective, inadequate resources.  What’s more, neither they nor anyone else know how well they are performing.

In 2003, the Laming Report made a huge number of recommendations.  The body of their report made it clear that there was too much process (bureaucracy) and too little time with the problem families.  Performance was assessed by activity (paperwork) and not by outcomes; the body of the report called for that to be reversed.  But outcomes did not feature in the actual recommendations which, ironically, focused on increasing the bureaucracy and made matters worse.

In the eleven years since, government has recognised that performance should be measured by outcomes and some limited progress has been made, e.g. for adoption services.  But where child protection is concerned, we simply get lip service.  No outcome measures have been suggested, nor the specific metrics, nor how they should be gathered.  In short we have no idea what “success” in child protection would look like or which local authorities are doing better than others.

When, last month, I asked the Children’s Department Minister, Edward Timpson MP, about performance measurement in this area, the answer was that they had turned the whole matter over to Ofsted.  The rest of his letter discussed the bureaucracy involved but there was not a word about how performance should be assessed.

This is, of course, a cop out: Ofsted should measure performance against standards set by government, as it does for schools.  Government is responsible for specifying what it wants in return for our money.  How else can they know whether to spend more or less?

Both Ofted’s “Framework and evaluation schedule” (published this June) and Rotherham inspection (published last week) have many references to the importance of measuring outcomes but nothing about what outcomes should be desired, what the metrics should be nor how they could or should be collected.  One has to feel some sympathy for the Rotherham local authority for being chastised for failing to do something that no-one has explained, not even the Department responsible.

Our children may or may not be safe with Children’s Services.  The bigger question is whether they are safe with this government.

Well, yes Sir Simon, but how do we calculate this?

Simon Jenkins is reviving the notion that clever people like himself, those Great and the Good, can tell all of the rest of us how to live our lives. His particular example is supermarkets but it could be anything at all really, given the proclivities of some to tell other people what to do.

We went from that High Street thing, to supermarkets, to out of town supermarkets and perhaps now to online sales:

Land is Britain’s most precious resource. The point of planning is to economise its usefulness.

We’d argue a bit there, Britain’s most precious resource is Britons. Their, our, accumulated knowledge, labour and the accumulated labour (also known as capital) handed down from our forefathers. But that aside, yes, of course, we wish to create the maximum economic benefit from whatever resources we have (and that does not mean just money, of course not, we’re talking utility here).

At which point we’ve got to ponder, well, how do we do that maximisation? And the truth is no one knows. That’s why we cannot plan. Should someone, in the 1980s, when considering a planning application for a supermarket have predicted the rise of the internet, Amazon and Ocado? Could they have done so? In the 1990s?

Yes? No?

If not, then it couldn’t have been planned for, could it?

At present, smart planning ought to be thinking ahead of the boom in online shopping. What mistakes might there be in pandering to its gargantuan appetites? What are the implications of every street jammed with home delivery lorries? What of every suburb blighted with distribution centres, supplied by giant hangars littering every motorway?

The correct answer here is “we dunno”. Nor do you and nor does anyone else. We’re all just going to have to suck it and see. Or, as we might put that a little more formally, allow the market to sort it all out. We consumers will work out which of the various options we ourselves prefer, those who cater to our desires will prosper and we’ll end up with a system that might not exult entirely everyone but which does the best to provide aggregate human utility that can be managed at this stage of technological progress.

And yes, that does mean that Sir Simon and his ilk don’t get to plan it all for us. Exactly what annoys them all so much of course.