Friday, 18 October 2013

How (Repeating) Short Term Crises Can Create Permanent Market Effects

Today's post comes off the back of Scott Sumner's note in his blog that short-term crises may not only have temporary effects on markets:

"I am very sympathetic to the view that market setbacks in response to short-term crises are temporary. It certainly looks that way. However it’s a dangerously misleading way of looking at markets."

I think Scott is right and I think the complication comes when we stop looking at one-off short term crises and instead start to look at repeating short-term crises, like the current budget rounds. In these latter cases we should expect to see long-term effects on markets up until the budget issues are finally resolved.

I see this effect being based on three factors: (i) market attention to/awareness of the risk (if they can't see it, they can't price it; if they don't care about it, they don't price it); (ii) the probability of the event occurring; and (iii) the magnitude of the event it is does occur.

Turning to (i) first – the market doesn't have an infinite supply of attention, rather like NASA scanning the sky for asteroids, it can only pay attention to a certain percentage of risks at any one time.   That means that if a risk isn't immediate, or one that is repeating, the market has a habit of forgetting about it.

This is what happened with the most recent US government shutdown.  The last budget round set a clear precedent that this budget round would lead to another crisis that wouldn’t be resolved until the eleventh hour.  But budget crises hadn't yet formed a pattern, like those set by Europe last year where there was a continuous market focus on Italian debt auctions, and as a result, the market didn't pay attention to the latest budget round until it was right on top of it.

That is why markets began to slide on 20 September, a full 11 days before any government shutdown would begin, despite being more than aware from past form that resolution shouldn't be expected for at least a week.  Markets started to fall because they remembered the budget round was coming, not because probabilities had altered.

The difference going forwards following this latest budget round is that budget crises have now begun to form a pattern.  Indeed we know that the next battles are only three months away (one in January and one in February) and that isn't long enough for the market to forget them.   This awareness will mean that markets will continue to price budget crises into their models and thus affect pricing until the repeating cycle is broken.

Turning to (ii) – the fact that the last budget round was a crisis does not in and of itself make it any more likely that the next budget round will similarly be a crisis (but it does make the market more aware to them).  What matters is how that last crisis ended – was there final resolution or do the terms of the deal leave matters unresolved.

Its uncontroversial to say that the deal struck to end the shutdown wasn't a home-run; it didn’t forbid budget-ransoming in the future, nor did it deal with the issues that made Republicans hold a gun to the head of government.  That said I think it's fair to say the will of the Republicans has taken a hit and the next round should be easier (although not all agree).

As a result I don't see the probability of the next budget round ending in crisis as being as great as it was heading into this budget round. However it's not hard to see a situation where Republicans had managed to keep public opinion more on their side and be fired up for January.  In that circumstance the market would see a higher probability of budget crises next round coupled with an increase awareness of budget crises.

Together that would mean a long term depression of market expectations until either (i) the probability of crises decreased; or (ii) the cycle was broken.

On (iii) – the magnitude of a US default is something that I don't think the market can price accurately for the simple fact that a global reserve currency doesn’t default that often and so there isn't the data to make an informed estimate.  On top of that, we couldn’t calculate what would happen when Lehmans went down so I doubt we can make anything like a fair guess for a US default.

As a result I don't think that the events of the last few weeks have moved the markets view on the magnitude of the crises, if anything I worry that having walked up to the precipice there will be slightly less worry next time (quite wrongly.)

To wrap up – I'm not sure that this budget crisis has put a long term dint into the markets, primarily because the Republicans lost so much political capital. However, through the mechanisms I outlined above I think that repeating short-term crises can cause long term effects and we're just lucky that this one (probably) hasn’t.

Wednesday, 16 October 2013

ONS Labour figures good - Vacancy Rate back on trend

The ONS today released the employment figures and the headline figures remain encouraging (

Seasonally adjusted unemployment for 16-64 year olds stands at 7.9%, 0.2% down on last year and a -0.1% change on the quarter.  There's also been a big jump in full time employment (3% YOY, 1.0% on the quarter) and a fall in part-time employment (1% YOY, 0.2% on the quarter) which, along with increased employment numbers implies that people may be moving back towards full time employment.

But I think the best indicator (and source of the happiest news) remains to be the vacancy rate.  For 17 out of the last 20 reporting periods the number of vacancies has been increasing and this was again the case for Jul-Sept (+6,000, 1.2% change on last quarter and 12.5 YOY).


There are now 2.0 vacancies per 100 jobs (compared to 2.3-2.5 pre crash and 1.6 in late 2009). After a slight shudder with the last report, its good to see the numbers back on trend although I'm going to predict it won't be until around Q1/2 2015 that we see vacancy numbers back to pre-crash rates.

For any American readers I treat the vacancy rate as a British version of your quit rate that the BLS ( out.  The argument for the quit rate being a good indicator is that people only quit their jobs when they have positive expectations of the labour market.  The vacancy rate doesn't quite capture the optimism (or lack thereof) of workers but remains a decent proxy.

Monday, 14 October 2013

Textbooks and 21st Century Markets

With the start of a new academic year, I've been in the market for textbooks and have been struck by a couple of things.


First, why does every course specify two textbooks, one of which can only be purchased for £50 with no second-hand copies retailing for less than £40 and at the same time suggest another textbook that can be found for £10 second-hand?  And who buys the first book?


Second, as Joe Kraus, the dotcom pioneer said, whilst the 20th Century was about dozens of markets of millions of consumers; the 21st Century is about millions of markets of dozens of consumers. My anecdotal evidence for this? I was looking to buy the first-year econ textbook Lipsey and Chrystal's "Economics, 12th ed." and I found that the cheapest place to buy it was on the Amazon-like Abebooks.  When I bought the book I didn't really look at where the retailer was based (I actually thought it was the US) but when I logged into my DHL tracking page I noticed that my seller was located in Delhi, India; and that got me thinking, how big is my Indian booksellers market for the Lipsey book I just bought? Is he a 21stcentury Krausian retailer?


Time for some market-sizing. In given year there are roughly 21,000 economics and politics students in the UK (  Let's assume that only half of courses specify Lipsey as one of the recommended texts.  And let's also assume that the majority of these people want to buy the cheap Lipsey book rather than the more expensive option and finally let's assume that the number of students from other disciplines that want the Lipsey book compensates exactly for the econ/politics students that don't want the Lipsey book (I know this isn't precise but heck its ballpark). 


Of those 10,500 students on Lipsey-recommending courses roughly 3,500 will be first years who would be interested in the book.  Extrapolating for the English-speaking student body around the world, we can estimate that the total market for the book is going to be around 50,000 individuals. 


But that isn't the whole picture - according to this survey ( only 23.4% of students buy the required textbooks and of those 59.4% buy secondhand books.  Now I think that (a) the students that answered 'no' and put themselves in the 23.4% assumed the question to be asking if they bought all required books rather than just some and (b) Lipsey is a mainstream and relatively important book; as a result I'm going to be more hopeful of my contemporaries and say at least 60% bought it. 


From all of that we get the second-hand market for Lipsey as being 50,000 x 60% x 59.4 %; i.e. somewhere around 17,820.


That is a tiny number of people but that still isn't the tiny market that potentially exists for online second-hand retailing. Why not? Well from experience I expect that a lot of undergrads buy their second-hand books in person either from a store or by inheriting from older students.  Therefore the number of people looking to buy this book online could be as few as 5,000.


This is still more than a dozen but it is exactly the sort of market that Kraus was talking about.  My retailer can hold second-hand first-year English-language textbooks for a range of subjects and know that because of Abebooks and the internet he can reach a market to whom he can sell those books.  That is the internet economy.


p.s. One last thought – how amazing is international shipping? My order was with a bookseller in India who then shipped my book via DHL. It has since taken a trip to Liepzig, Germany, gone to the East Midlands and is now coming to me via Heathrow. All that for £3.69! That’s the price of a meal deal at Boots.

Friday, 30 August 2013

Syria : We must intervene

I have no preference as to who should govern in Syria. It is not my place, it is not our place, to decide for a people who should lead them and any questions of British self-interest should not and cannot influence our thoughts.

However, there comes a time when the issue we are faced with is not one of sovereignty, it is one of protecting humanity. On the intelligence we have, or at least that we are informed of, the Assad regime has begun to use chemical weapons against civilians. He is massacring his people. This isn’t hypothetical; its real and its happening right now. There can be absolutely no doubt that this crosses a red line and as a nation that believes in the rule of law and the sanctity of human life we cannot ignore it.

I know that our nation bears the scars and the burdens of our recent commitments to Iraq and Afghanistan. I know that parents have lost sons and children have lost their mothers. We have grieved for them, and we remember them. Every day. However we cannot withdraw from the world because it has got too rough, because the risks have got too high. When we are left with no option we must remain willing and committed to the use of force.

The vote yesterday showed us that the nation is torn and we are afraid of becoming embroiled in a battle that does not have an end point or an immediate impact on our daily lives. I understand those concerns and I share them. But I also share our nations belief in basic laws of man. The part of us that says we will not allow a bully to butcher his people. The part of us that compels us to defend those who cannot defend themselves. And right now those brave parts of us have to override our fears.

I don't want war. I don't want our forces to be in harms way. But the vote yesterday was wrong. Civilians are being massacred with weapons that have long been banned by international law. Our members of parliament must be led by their consciences and decide whether their fears matter more to them than the lives of the civilians of Syria.

I pray that they overcome those fears, anything else is cowardice and a betrayal of our humanity.

Thursday, 3 January 2013

Income Tax Volatility

Evan Soltas has written a number of pieces on income tax/deficit volatility that are excellent but I think his conclusions on income tax get away from him a bit as there is a larger picture to consider.

When setting a tax code, the primary concern of government is not tax receipt volatility but rather the generation of tax incomes and the effect of that taxation on the economy. Tax receipt volatility is a concern but is a secondary one, and is not a concern during boom years (as there is no volatility).

Once that frame of reference is placed on the question I think we see a slightly different issue. Let's play it out.

We'll assume that there is high pre-tax income equality (which also happens to be true for Evan in the US and me in the UK). And we'll also assume that the government is ignoring my two main concerns and focusing squarely on minimising tax receipt volatility. Applying Evan's analyis, it appears they're going to do this by levelling out income tax rates (we're going to come back to this).

So the secondary issue of income tax volatility is resolved, but what about my two "major" concerns - total tax receipts and effect on the economy?

If we first assume that the government wishes to maintain the same level of revenue following a levelling out of the income tax rates, we're either going to see: (i) a reduction on the rate of taxation on high earners and and increase on low and mid-income earners; or (ii) a reduction on taxation on high earners + some other taxation; or (iii) some mixture of the two.

On (i) that's going to hit consumption of those two groups pretty hard, as to rebalance the share of income tax revenues generated by high earners is going to take a far greater increase in taxation on low and mid-income earners.

On (ii) I'm not sure what taxation would be appropriate? I expect consumption to be similarly volatile to income for high income earners (and untargetted consumption taxes are regressive) and corporation tax/cap gains isn't going to help as these two figures are just as volatile as high-earner incomes.

Because I don't see (ii) being effective, I don't see (iii) working out either, as it ends up just being (i).

My feeling then is that to not cause unwanted effects of such a change in consumption economy, you're going to have to accept less tax revenue as a trade off for less volatility.  If tax revenues fall and we don't see a spending adjustment (and if anything I'd advocate an increase in current spending to boost demand), the deficit increases. 

I'm not a deficit hawk, in fact I don't think government debt is a particularly major issue in the UK or US right now - especially compared to the severity of the threat posed by the recession. And for that reason I don't mind the deficit swelling during the recession.  But the only reason I'm happy with it is because I know that if I keep my volatile tax code the same (bar any reduction to stimulate growth), when we hit growth again, those same high earners that are not paying up right now, will return to their normal levels.  But if you don't have that uplift when growth returns, I'm not sure how Evan is going to get rid of the deficit he's growing.

In fact, when the economy is experiencing sustained growth, the government is (or should be) aware that income tax revenues will be volatile if we enter recession. Armed with that knowledge, the question is not tax policy but spending policy. The government knows it will have less revenue at some point in the future and so should hold back spending during growth years - it should run a surplus. A failure to run a surplus is what causes the problems with volatility, not the volatility itself.

All of this is said with (a) a lot of respect for Evan's excellent blog and (b) a lot of humility as someone who is equally a student of the subject.