Tuesday, 18 April 2017

Why regulation is hard

http://marginalrevolution.com/marginalrevolution/2017/04/the-spirit-of-the-law.html

The tyranny of the UK two party system

Since the General Election is called today, here's why doing the maths is important.  The BBC news website, asks: with Labour a long way behind in the polls, will there be a landslide of wining seats for the Conservatives?  The answer is not nearly as much as you think, for the following interesting reason.

BBC links to a a study published by the Fabian Society, making the point that "Labour is too weak to win, and too strong to die". By this they mean that Labour has, basically, a very large number of safe seats with large majorities.  So if it loses votes, it simply has a lower majority in each of its seats, with relatively few loses ("too strong to die").  But it is also "too weak to win", in their words, since a small swing towards it just raises its majority in seats it already will win.

 Figure 4 contains the key data, all using national shares of the vote.



Columns 1 and 2 show what happens if Lab get 20% of the vote, a very very low share, Con 45 and the others are split.  Lab doesn't win, but still gets 140 seats (currently it has 232).  Columns 3 and 4 show what happens if the Libs get up to the same as Labour, at 20.  Note that Labour hardly suffers: indeed it wins more seats and the Libs get hardlly any more.  Columns 5 and 6 show the same.

The BBC website summarises this nicely:


Yet the figures suggest that despite the current low standing, a large majority of Labour MPs would be re-elected.

That's because a lot of them are in safe seats.

An interesting feature of recent British electoral politics is that the number of safe seats has increased whereas the number of marginal seats has fallen.

In 1992, 169 of the seats in Great Britain were won with majorities under 10% - a common definition of a marginal seat. In 2015, that number was just 119. And only 49 of those were won by Labour.
On the other hand, the number of very safe seats has increased. In 1992, there were 155 seats won with majorities above 30%. In 2015, there were 223.
It continues:
A further interesting quirk of British elections is that the extremely safe seats in the country tend to be Labour seats, even when it loses the national vote.

The 11 safest seats in the country are all Labour - as are 17 of the safest 20.

That doesn't mean those constituencies could never elect an MP from another party. In 2015, Labour lost some very safe seats in Scotland as a result of the SNP landslide. But it does mean that for many Labour MPs, even the current polls shouldn't cause them too many worries about their own futures.

This trend became even more acute in 2015.
The Conservatives pulled off their surprise victory by winning votes just where they needed them, such as in Liberal Democrat-held seats in the south west of England - which is now almost entirely blue.
On the other hand, Labour piled up lots of extra votes in seats it already held comfortably.So nationally their vote share went up even as they lost seats overall.

I think this also answers another question: why didn't Lab MPs, who don't like Mr Corbyn very much by all accounts, but cannot get rid of him, try to start another party?  Part of the answer is that a core group of them, on these maths at least, don't have to worry about their seats.

Friday, 17 March 2017

House price to earnings ratio in London, 1969-2014

This is remarkable....




Figure 9 of Working Paper 72
House prices in London – an economic analysis of London’s housing market
Joel Marsden
November 2015

Wednesday, 10 August 2016

Common Cognitive Biases

A number of bloggers, such as Chris Dillow or writers such as Danny Finkelstein write insightful stuff which blends economics, politics with cognitive biases.  I found the following table of commonly studied cognitive biases from Wikipedia very helpful as a summary:

Name

Description
Fundamental attribution error (FAE) Also known as the correspondence bias (Baumeister & Bushman, 2010) is the tendency for people to over-emphasize personality-based explanations for behaviours observed in others. At the same time, individuals under-emphasize the role and power of situational influences on the same behaviour.
Confirmation bias The tendency to search for or interpret information in a way that confirms one's preconceptions. In addition, individuals may discredit information that does not support their views.[29] The confirmation bias is related to the concept of cognitive dissonance. Whereby, individuals may reduce inconsistency by searching for information which re-confirms their views (Jermias, 2001, p. 146).[30]
Self-serving bias The tendency to claim more responsibility for successes than failures. It may also manifest itself as a tendency for people to evaluate ambiguous information in a way beneficial to their interests.
Belief bias When one's evaluation of the logical strength of an argument is biased by their belief in the truth or falsity of the conclusion.
Framing Using a too-narrow approach and description of the situation or issue.
Hindsight bias Sometimes called the "I-knew-it-all-along" effect, is the inclination to see past events as being predictable.




Thursday, 28 July 2016

More on how regulation can get in the way of competition

A follow up to http://haskelecon.blogspot.co.uk/2015/06/the-too-big-to-fail-subsidy-in-uk.html.


The CMA's inquiry into banking is coming to an end.

The challenger banks have written, Letter from 8 challenger banks, to say the main problem is their disadvantaged funding, via a regulatory formula that means they have to set aside much more capital than a large bank, something they thing is tied up in the Too Big to Fail Category. They say



In reply, Letter from Professor Alasdair Smith to 8 challenger banks, the CMA say two things.

First, "it is an important provisional finding of the CMA investigation that the funding advantages of the incumbent banks derive in large part from the inertia in the customer base. I believe that the remedies package we have put forward will create a more fluid market"

But second, in response to the regulation question they delegate it to others.

"In our view the issue should be taken forward by the bodies which have primary responsibility for the safety and soundness of the banking system"

and
" It would be in no one’s interests to create ambiguity about regulatory responsibility in this area in particular given the international context in which reform will be required. This is also the case for addressing the ‘Too Big To Fail’ issues you raise, which are being tackled by the Bank of England and the Financial Stability Board’s reforms to the resolution regime for global systemically important banks."

I had hoped the CMA inquiry was going to be a chance to address the Too Big To Fail issue. It would seem not.

(See also this exchange of letters with Sir John Vickers, Sir John Vickers (11.7.16) CMA response to Sir John Vickers (11.7.16)).








Examples of how basic research affects real products

From Gianni de Fraja, paper in the current Rand Journal of Economics
Optimal public funding for research: a theoretical analysis (pages 498–528)

Nelson's early work (1959) already reports many examples of basic research projects that illustrate these characteristics. Among the cases studied more recently, Moody (1995) describes in detail the numerous strands of basic research which allowed the creation of the ubiquitous CD. A Global Positioning System (GPS) navigation system would be far too inaccurate to be of any practical use without corrections of gravitational effects central to the theory of relativity (Haustein, 2009). The abstract mathematical problem of covering a surface with tiles lies at the foundation of our understanding and exploitation of superconductors (Edelson, 1992). Gauss's investigation into the distribution of prime numbers has led, with the contributions of some of the best mathematical minds over the course of two centuries, to the possibility of unbreakable cryptographic codes, without which e-commerce would not be possible (du Satoy, 2003).10