Wednesday, 31 October 2012

Various teaching links: How to know more than the next guy.

There’s an old joke about two men in a forest trying to get away from a tiger.  One puts on an old pair of running shoes.  “You can’t go very fast in those” says the other.  “I only have to go faster than you” says the first. 

Regrettably,  much of life is being faster than the next guy.  So here’s a lesson in how to get ahead, drawn from the report into the West Coast Mainline bidding fiasco, which I just saw here.

As you remember, the bidding process collapsed in ignominy and will have to be rerun following the admission of errors in the process by the Department for Transport (DfT) who conducted the bids.  The inquiry into the fiasco has a preliminary report out.  I read the following lessons.

1.      The bidding works like this.  The winner of the bid gets all the revenues from the line they operate.  But, they have to pay a per year fee to the DfT to operate the line.  The DfT recognizes, correctly, that revenues might rise or fall depending on general economic activity which cannot be foreseen.  Thus there is an adjustment formula that adjusts the fee in line with GDP, in particular reducing it if GDP falls.  So you can win the bid by offering a large fee, knowing that the fee will fall if unexpected bad times come along.
2.      At the same time, the DfT does not want the franchiesees to go bust.  So you can also win the bid by offering to hold a lump sum of money, which payable to the DfT in bad times.
3.      What’s key for the bidders is to know whether they can win by offering:
a.       to hold a large lump sum, but bid a low fee
b.      bid a large fee, but hold a small lump sum
4.      The DfT have an economic model, which tells them the answer i.e. if  a bidder decides to adjust the lump sum, how much they can vary the fee..   Here’s what went wrong.

a.       they were unwilling to show the model to the bidders who were just given a number telling them the trade off between lump sum and fee.   As it turned out, that number was not based on the model at all, but on some other procedure kept secret from the bidders.
b.      The model worked out payments in real terms, in 2010 prices.  But the DfT thought the model results were in nominal terms.  So bidders were given a figure for what they were told was an adjustment in nominal terms (see paras 5.14.3).  This gave them the wrong price.  This matters since the bids last for five years.   So if you are told to make an adjustment of £X in 2015 and that mistakenly in real terms, it can drastically understate the correct nominal figure (by the compounded price change over 5 years).  As the report says 5.15. “Had they been converted into the nominal terms, which they should have been, significantly increased [adjustments] would have been required”

So here’s how to get ahead by knowing more than the next guy.

  1. A nominal number is in pounds, e.g. my salary was £400 per week last year and £420 this year, a rise of 20/400=5%
  2. A price index tells you how much the average basket of goods costs from year to year e.g. the basket costs £200 last year and £210 this year, a rise of 10/200=5%.  
  3. Since wage and prices have risen by the same, the “real” wage is unchanged.  Since a numbers in real terms is the nominal number divided by the price index and index of the real wage is £400 last year and £400 this year.

Tuesday, 30 October 2012

Various teaching links

1. Distinctive capabilities.
A marvellous example from the FT today using Kay's capability analysis on the failure of the London black cab manufacturers Manganese Bronze.  Their two capabilities: regulation and reputation, ran out. Interesting articles here, pointing out for example:

"2007 was the last year the company turned a profit. While Manganese Bronze had updated its cabs over the years, they were still being built on a basic structure that traced its heritage back to the first black taxi, dating from 1948.  Despite its outdated product, the company enjoyed a protected market because of a rule that London taxis have a 28ft turning circle.

But by 2008, the TX4 faced serious competition from Mercedes-Benz’s Vito, which met the 28ft circle rule but was more fuel-efficient and cheaper to run.  In just over four years, the car has captured 38 per cent of the London market"

2.  Adjustment along many margins
In our supply and demand models, price and quantity adjust to restore equilibrium.  But what if they, due to regulation for example, cannot adjust and are not at the equilibruim point?   The lessson of economics is that in such a market there is an opportunity to trade.  So its likely that there will be some other form of adjustment.  Here's a great example from Tim Taylor on rent control.

Monday, 22 October 2012

Is British Growth Held Back Due to Small Cities?

The Economist thinks so: larger cities would raise learning, communication of ideas etc.

The SERC is not so sure:

"City size and diversity, however, provide an economic payoff: a critical mass of people, resources and ideas help produce agglomeration economies
(Glaeser 2011). Increasing that critical mass helps raise productivity, therefore: the consensus
from recent studies is that doubling employment in a city raises average labour productivity by
around six percent, although these effects are much more important for some types of economic
activity(Melo, Graham et al. 2009). They are much more important in precisely those sectors of
economic activity in which the British economy is specialised and our most prosperous cities – the Londons, Cambridges and Oxfords – are particularly specialised: skill intensive traded services.
 Although urban density is strongly correlated with the effective or functional size of a city there is no evidence that density itself is a cause of these observed agglomeration economies. It seems more likely that density is the outcome of agglomeration economies as both households and firms bid up the price of land to benefit from them thus causing development to be at higher density. Indeed Cheshire and Magrini (2009) find that once all other factors including city size are controlled for, higher density is associated with slower urban economic growth."

And they have interesting examples:

Two examples illustrate the difficulty of separating out density effects. 1) Building CrossRail, for example, will likely reduce the density of the London region as a whole as people take advantage of quicker travel to
move out to cheaper land. But it will still increase the effective size of London since with easier travel the costs of productive interactions between economic agents will fall and their potential number will increase. 2) Take two cities with identical populations and borders: building more houses will increase density. But it is then hard to attribute any subsequent economic changes to higher density, since population size has also gone up.

Saturday, 20 October 2012

How the UK can recover: Nick Crafts has the answer

Nick Crafts gave the RES policy lecture on 18 October.  He's right on almost everything: here's what he has to say

If fiscal consolidation continues and radical changes to monetary policy are ruled out, it is mainly ‘supply-side’ reform that can restart UK growth without doing longer-term damage to the economy. Among other things, that means repairing infrastructure, improving education, reforming taxation and tackling the restrictive planning system.

And here's the answer

But one area that could deliver both short-term stimulus and long-term efficiency is private house-building – as happened in the 1930s recovery from recession. Today’s planning restrictions mean that the stock of houses is three million below and real prices are 35% above what they would be if market forces operated freely.

Slides of Nick Crafts’ RES Policy Lecture.

Various teaching links: agricultural productivity and patents

The fantastic Tim Taylor has two very good posts of relevance to my Economics of Innovation students.

1. Can Agricultural Productivity Keep Growing? 
The reference is to the Fuglie et al paper, which I have referred to in the slides.   They  have the following fascintating graph, showing that Agricultural productivity has been more and more dominated by TFP growth in recent years. And that TFP is strongly correlated with country R&D.  So sub-Saharn Africa, which needs the TFP, might not get it if its not doing any R&D. 

 2. Patents Tipping Too Far: Three Examples

His post explores the following.
The basic economics of patents as taught in every intro econ class is a balancing act: On one side, patents provide an incentive for innovation, by giving innovators a temporary monopoly over the use of their invention. This temporary monopoly rewards innovation by allowing the inventor to charge higher prices, and thus the tradeoff is that consumers temporarily pay more--although consumers of course also benefit from the existence of the innovation. Like any balancing act, patents can tip too far in one direction or the other. On one side, patents can fail to provide providing sufficient incentive (that is, large enough profits) for inventors. But on the other side, patent protection that is too long or too rigid can lock profits for early innovators for an extended period, both at the long-term expense of consumers and also in a way that can cut off possibilities for future innovators.

The Economic Way of Thinking

Economists think about things in ways that involve words like “Value”, “incentives” and “utilities”.  Such words are often used by other subjects, so here is an attempt to explain WIHIH (what in hell is happening).  Mostly it is, I think, using the same word for different ideas.

  1.  Value

This is a source of endless confusion.  Here’s the Economist’s notion. 
a.       Think of a demand curve.   That curve traces out the willingness to pay for an additional item of the good.  If consumers are different, it’s the cumulative willingness to pay (WTP) starting with the person who values the good most, Nick, Harriet and Ali in the coffee example.  If consumers are the same, its the WTP arising from diminishing marginal utility of the good.
b.      Now suppose the price is set, by supply or law of something else: say £2.00.  At a price of £2 the Economist says, that gives the willingness to pay of the marginal consumers.  It does not give the WTP of the inframarginal consumers, indeed they earn some consumer surplus precisely because they are not marginal and it’s the marginal consumer’s WTP that the price reveals. 
c.       So what can we say about value in the marketplace?  To economists, all we can say is the value placed on the good by the marginal consumer, here £2.  All the  other consumers who are buying get consumer surplus.  We might, if we knew consumer surplus be able to work out the total WTP as well as the marginal WTP.  And if we wanted to call it that, we might be able to call total WTP total value.  But in Economics we tend not to do that; we reserve value for “value added” which is an accounting term (revenue less cost of intermediate goods used up in production).  Even if we did, economists don’t tend to talk about value in the market place since we would have to be able to count up consumer surplus to measure it.  At best, prices reveal the marginal customer’s WTP, which you might call how much the marginal customers’ values the good.  We need to know a lot more if we want to talk about every customer’s WTP or, if you so define it, their value.
d.      The emphasis on prices as revealing marginal values is, I think, helpful, since it avoids the endless debate on what is “value”.  Diamonds are very expensive and (in Western countries) water is very cheap.  But diamonds are useless baubles whereas water is vital for life.  We can then argue about whether diamond are more or less “valuable” than water, but the Economics approach cuts through this.  All we need realize is that if the price of Diamonds are high, that suggests the WTP, or “value” if you like, of an extra Diamond is by the marginal diamond buyer is very high.  Whereas with a lot of water around, the WTP, or “value” of an extra bottle of water is very low.  I think we can agree on marginal values and we can just avoid having to agree total values.
2.   Prices as incentives

Some object to prices being an incentive.  There are a number of objections.
a.       Consumers/firms don’t care about prices.   Economists: fine, that just says elasticities are low. 
A related notion is that you need, as a business, to emphasize quality, customer loyalty etc. and this somehow is more profitable.  The Economists' answer is that such things don’t come for free.  Getting loyal customers presumably means advertising, R&D etc. all of which costs.  It might shift out the demand curve, or make it less elastic, but its still not a guarantee that it will be profitable: that depends on the discounted benefits to incurring such costs.  Economists are, IMHO, good at looking at correlations between, say advertising spend and future revenue.  But the Black Box of just what neurones in the brain are activated by advertising is not well studied (for more on intrinsic motivation, see below).
b.      When Consumers/firms use prices to incentivise others, this might lead things that are not intended.  Example.  The firm rewards its workers for working quickly i.e. sets a good “price”, in this case a wage, for fast work.  But, then workers produce poor quality.  Economists’ answer.  But this certainly means that economic actors respond to prices! All that is happening here is that they have a menu of actions of dimension N, but the menu of prices they face is less than N, say M. 
c.       A more subtle criticism.  Consumers/firms/economic actors respond to “intrinsic motivation” e.g. fairness, responsibility.  If prices are then used, that will “crowd out” such responsibility.  Example: setting rewards for good performance at work is not a good idea because it offends people’s sense of responsibility and makes them perform worse (a reward for good performance improves behaviour due to extrinsic motivation, but worsens it since it weakens intrinsic motivation, perhaps because it signals that the employer does not trust the worker or if the worker tries due to a concern for social status which is undermined when it is paid for).  This is a key idea in Michael Sandel’s recent work and is discussed in a very interesting recent article by Gneezy, Meier, Rey-Biel, Journal of Economic Perspectives, Fall 2011.

The famous example is Titmuss (1970), “who argued that paying people to donate blood broke established social norms about voluntary contribution and could result in a reduction of established social norms about voluntary contribution and could result in a reduction of the fraction of people who wish to donate.the fraction of people who wish to donate”.
This is very controversial in education, with many programs now starting to pay High School students money to attend school, do exams, hand in work etc.
So what do we know about this area?  My take on the piece is that it is a bit horses for courses. In some contexts intrinsic incentives are so weak that financial rewards are very helpful.   “The current evidence on the effects of financial incentives in education The current evidence on the effects of financial incentives in education indicates moderate short-run positive effects on some subgroups of students, at indicates moderate short-run positive effects on some subgroups of students, at least while the incentives are in place.”
But the article also points out that incentive programmes must be clearly designed to be sure that such crowding out does not occur.  The conclusions are very interesting:

“When explicit incentives seek to change behavior in areas like education, contributions to public goods, and forming habits, a potential confl ict arises between the direct extrinsic effect of the incentives and how these incentives can crowd the intrinsic motivations in the short run and the long run. In education, such incentives seem to have moderate success when the incentives are well-specifi ed and well-targeted (“read these books” rather than “read books”)…

In encouraging contributions to public goods, one must be very careful when designing the incentives to prevent adverse changes in social norms, image concerns, or trust.  

Incentives to modify behavior can in some medial cases be cost effective. The medical and health economics literature intensely investigates whether, and when, prevention is cheaper than treatment (for example, Russell, 1986).  The question is economic rather than moral: certain prevention activities can cost more than than they save,  For example, cholesterol-reducing drugs can cost hundreds of dollars a month; simple exercising could, in some borderline cases, replace these drugs.
Our message is that when economists discuss incentives, they should broaden their focus.  the effects of incentives depend on how they are designed, the form in which they are (especially monetary or nonmonetary), how they interact with intrinsic motivations (especially monetary or nonmonetary), and what happens after they are withdrawn.”

Update on the UK recession, long and large, or short and long

Chart 5: UK output per worker compared with pre-crisis trend, seasonally adjusted

The ONS released a new piece today on the UK Productivity Puzzle.  Some details:

1. current UK output per hour compared with other recessions is a disaster (output per worker does not look much better)

2.    That makes output per worker 10% lower than if it had continued on its pre-recession path:
Chart 4: UK productivity levels, output per hour during UK recessions, seasonally adjusted

3.   The international comparison for the UK is not good either:

Chart 6: International comparisons of output per hour productivity growth since 2008 Q1, seasonally adjusted

Chart 6: International comparisons of output per hour productivity growth since 2008 Q1, seasonally adjustedh

As the graph shows
1. everyone is worse than the US
2. the UK fell initially by less, but then fell back in the 2011 period, so that it is now at the bottom of the pile relative to 2008Q1 at least. 

As Peter Pattinson said at the seminar launching this document, the early productivity period is not so much of a puzzle, since barring the US, everyone seemed the same.  What is very mysterious is the post 2011 performance.

Could the GDP data be wrong?  Walton and Brown look at this.  Two crucial graphs:

1. mean revisions have got smaller since the 1960s and are negative in recent years (chart 3)
2. since 2005 the revision size is smaller than in the 1990s at 0.3% pa to growth.  So a revision, if upwards would add on average this.  Not enough to restore the 10% gap, but might be enough to restore the UK relative to other non-US in the chart just above.

US recessions and the current US recovery

I wrote about the recession here for the UK. In the US, Krugman and Taylor are having an argument:

1. Taylor says the US recovery is very weak relative to that from other financial crises, and blames policy.
2. Krugman says this is politics and the US recession is just what you would expect from a financial crisis.

Reinhardt and Rogoff have this graph:

Reinhart: Fig 1

There are two questions:
Q1.  is the US recovery, starting from the bottom of the cycle, currently slower than the 1930s?
Looking at the slope of the 2007 line starting from the bottom of the V, it grows slower than the wide dotted line, but about the same/faster as the small dotted line.   As Jim Hess has correctly pointed out to me, correcting an earlier mistake, the wide dotted line excludes the 1930s, and the narrow one includes it.  So the 1930s must have been slower recovery and hence current recovery is faster.  Score one for Krugman.

Q2.  are US recessions, starting from the peak, longer to get back to the peak, when they are financial crises?  Yes they are, according to Reinhardt/Rogoff  comparing different recessions according to type.  Score one for them and Krugman.

So its two different questions being compared.

Gettting the housing market going? Using simple economics for policy analysis

Ed Balls says let's use prospective 4G licence fees to fund a stamp duty holiday.  On this idea, we know what will happen since we have some evidence from when it was tried before: Chris Giles points us to the paper. from HMRC that evaluates the effect of a temporary relief on stamp duty for first time buyers March 2010-March 2012.

As my Imperial MBA students will know, the effect of a tax reduction in a Supply and Demand model depends on the elasticities of demand and supply.  Suppose as seems very likely due to planning, the supply curve is very inelastic.   Then tax falls have minimal impact on quantity.  If demand curves are quite elastic then any falls in taxes translate into minimal falls in prices for buyers, but rises in prices for sellers. Finally, falls in taxes are expensive for the state, since with little extra quantity the gain in revenue from the additional marginal quantity is small, but the loss from the inframarginal is large. 

So we can use some very simple economics to make some sharp predictions.  

And the findings in the real world from the study?  They show Economics right on the money.  Here's an extract from the summary.

The effect on quantity is tiny: 
The number of additional transactions is therefore estimated to be closer to 0‐1  per cent (1,000 additional transactions) 
The effect on buyer prices is tiny:

Considering the impact on prices, ... This implies that the majority of the 1 per cent tax  relief was capitalised in higher prices. It is equivalent to stating that the post‐tax
outlay for buying property is estimated to have decreased by 0.3‐0.5 percentage  points. The relief therefore appears to have had a small impact on reducing the
overall outlay of buying a first home. 
The costs to the state are huge:

The cost of the tax relief was around £150 million in 2010/11..[with].. 1,000 additional transactions across the 13‐month period April 2010‐April 2011, this represents an estimated cost to the Exchequer  of approximately £160,000 per additional transaction in tax relief.