Bruce Greig

Bruce Greig, ADR Group Accredited Civil & Commercial Mediator

2020-10-06T09:18:00.006+01:00

During lockdown, I trained up as a Civil & Commercial Mediator with ADR Group

Mediation is all about finding ways to settle a dispute. To find common ground where seemingly there was none. Over the years, I’ve often found myself doing this in my businesses, for friends, and at the school where I am a governor.

I really enjoy it. Navigating my way to a negotiated agreement is pretty much my favourite work activity, neck-and-neck with building geeky spreadsheets.

It turns out that people might actually pay me to do this kind of thing, which would be lovely as a sideline or, even, a full-time job.

More on my new Bruce Greig Mediator website

Little side project to help schools run staff surveys (SchoolStaffSurveys.com)

2018-11-05T10:57:00.000+00:00

I've been a school governor since 2011. I was Chair of Governors for much of that time, and have seen two Ofsted inspections and four headteachers.

Trying to understand a school as a governor can often be like the blind men appraising an elephant. You get little glimpses of information here and there, which might lead you to confidently reach some conclusion or other, but often you just don't really know.

One thing which really helps, in my experience, was a detailed, robust and anonymous staff survey. We just used Google Forms to do this at our school, but it is a bit clunky and time-consuming to set up in Google Forms. So I built this simple little tool to allow school leaders to run a ready-made survey quickly and easily:

www.schoolstaffsurveys.com

There's a thee minute overview video below:

School Staff Surveys Demo Video from School Staff Surveys on Vimeo.

If you know any school leaders, do let them know.

Ten billion pounds of dodgy accounting

2016-03-26T08:52:00.001+00:00

**************
Update 23 Nov 2016:

The ONS, in October 2016, took steps to stop one of these scams working, the corporation tax one. Most accounting systems work on a accrual basis: you account for revenue and costs in the period they relate to, not necessarily in the period you get or spend the cash. The Government used to account for Corporation Tax on a cash basis, so they could artificially move receipts into different periods by simply changing the rules about when companies had to pay their bills.

No longer.

The ONS now requires Corporation Tax to be accounted for in the period in which it accrues, so George Osborne's little trick of hauling future years money into 2019/20 to flatter that year's figures would not have worked. This would have knocked £6bn out of his forecast £10bn surplus. See box 4.2 of the OBR's November report:

**************

Here’s some more detail why I’m saying the Chancellor is deploying Enron-like tricks to artificially flatter the Government's borrowing forecast for 2019/20.

First, some background: 2019/20 is important because that’s the year when the Chancellor has said he will eliminate the deficit. That’s in the Charter for Budget Responsibility.

The impression is that he will gradually chip away at the deficit every year until he’s eliminated it entirely by 2019/20. But that’s not quite what he’s doing. As you’ll see, some of the measures he takes will actually make the deficit worse in earlier years, in order to push some tax revenue or spending cuts into the crucial 2019/20 year. It is all about 2019/20.

For every one of the next three years, the deficit is now a lot worse than it was expected to be as recently as the Autumn Statement in November. But 2019/20 still magically shows a £10bn surplus:

(Source: OBR, EFO March 2016, Table 4.34)

It maybe a little strong to describe this as “Enron-like”. Perhaps it is more “Tesco-like”. Tesco deliberately recognised revenues earlier than they should have done, and recognised costs later than they should have done, in order to flatter one year’s accounts and the expense of the next year. Then the next year they’d scramble to do the same kind of thing until it all unravelled. You can’t do that forever, although as Tesco showed you can do it for quite a long time.

Here are the three measures which I’m calling out as decidedly dodgy:

Accelerated tax payments for large corporations (£6bn)

This one is somewhat confusing because it is a measure which would originally have given a one-off boost to 2017/18 and 2018/19 but has now been delayed by two years so that the benefit is seen in 2019/20 and 2020/21. Ostensibly this is to “give companies more time to prepare”. But if you are cooking your books, why bother cooking the books in years which don’t matter much? Do the cooking in 2019/20, the one year you really, really have to show a surplus. So, in the 2016 Budget, the Chancellor moved this measure so it took effect in 2019/20.

Here’s how this one works:

At the moment, large corporations pay corporation tax in instalments with instalments falling seventh and tenth month of the current accounting period and in the first and fourth month of the following accounting period. So if your year-end is December 2016, you’d have to pay an instalment of tax on your estimated full-year profits in July and October 2016, and the rest in January and April 2017.

With the accelerated tax payments, you’d have to pay in March, June, September and December 2016. This in itself is quite a radical change: you have to pay tax on your profits before your year is even over. But for our purposes, the important point is that in the year this measure is first introduced the Treasury gets extra money (in that year only), as they get some of the previous year’s money and some of the current year’s money in the same year. This a temporary blip: it is as if your employer changed from paying you at the end of each month to at the beginning of each month. The first month that happened you’d get two paychecks in one week: one for the end of the previous month and another, the very next day, for the start of the following month. That would be a great week for your bank balance, but you aren’t actually being paid any more really.

This measure moves £6bn of tax into 2019/20 from later years. In the Budget it appears to move it from earlier years, but that’s because the measure was originally announced as something for 2017/18, but was then delayed in the Budget “to give companies more time to prepare”. Hence the strange wording of “defer bringing forward”:

Move capital spending out of 2019/20 to earlier years (£1.6bn)

The Chancellor is moving £1.6bn of capital spending originally earmarked for 2019/20 into 2017/18 and 2018/19. The official description is that the government wishes to “accelerate investment plans”. But the cynical reason might be: why would you spend money in 2019/20, the year you really need to show a surplus, if you can shift some of it into earlier years where you’d don’t need to show a surplus?

This improves the 2019/20 figures by £1.6bn:

This has happened in each of the last three budgets / statements: capital spending has moved quite a bit, but whatever happens in other years, it only got lower in 2019/20. As the OBR politely point out:

(p140 of the Economic and Fiscal Outlook March 2016)

Increasing pension contributions from public sector employers (£2bn)

This is pretty convoluted and I’m not totally sure I’ve understood it correctly. But this is what I think is going on:

Private pensions work by investing your pension contributions over your lifetime to build up a pot which is big enough to fund your pension when you retire. Public sector pensions don’t work like this. They are just paid for out of tax revenue. But public sector employers still have a notional charge which comes out of their budget as if they were putting contributions into a fund. There is no fund, these “contributions” just come out of the department’s budget and back to the Treasury.

The level of these contributions is set, though, as if they were going into a pension fund. And the calculation of the appropriate level involves using a discount rate: an estimate of the future value of money. If you are expecting your investments to grow less slowly, you would reduce this rate … so to generate a fixed pot at some date in the future you need to put in more money now if the rate goes down.

This discount rate has gone down, because the economy is forecast grow more slowly.

The effect of this is that public sector employers (the NHS, schools, etc.) need to send back more money to the Treasury for these notional pension contributions. But their overall budgets are not increasing to cover these extra pension costs. They will have to cut other things within their budgets to come up with these higher contributions. This is effectively a cut of £2bn per year from 2019/20 onwards, but it is not being presented like that.

This improves 2019/20 figures by £2bn:

Taken together, these three measures improve the 2019/20 figures by nearly £10bn.

How much surplus is the Chancellor is forecasting for 2019/20? About £10bn. Fancy that.

The disability thing looked fine to me. It's the Enron-style accounting tricks we should be worried about.

2016-03-22T22:12:00.001+00:00

Lots of people won’t like me saying this (and I’m happy to be corrected if I’ve got this wrong), but the changes to Personal Independence Payment (PIP, aka “disability benefit”) which appeared in last week's Budget and triggered Iain Duncan Smith’s resignation actually seemed pretty reasonable.

Short version: independent reviewer in 2014 suggested that the “aids and appliances” bit of the PIP might not be working as intended and asked DWP to look more carefully at it. This they duly did and proposed a small change to the scoring system, with the intention of excluding people whose disability is quite mild and does not create extra costs for them.

Long version: A third of PIP claimaints base their claim solely on their need to use “aids and appliances” in certain activities. Under the current scoring system, you can get the basic level of PIP if you need to use something like an easy-grip kitchen knife to prepare a meal (2 points), a grab rail to use the bath (2 points) and loo (2 points) and needed to sit down on a chair to get dressed (2 points). That lot adds up to the 8 points to get PIP. Mobility aids, incidentally, are nothing to do with this, they are scored separately.

The number of claims which rely solely on aids and appliances has tripled in 18 months. Even before that, the independent reviewer, Paul Gray, suggested back in 2014 that the aids and appliances bit might not be working correctly, although his review didn’t have access to the data to be sure. But he recommended that the DWP look into it. The DWP then looked at 400 claims which were based solely on aids and appliances. Of these, they judged that 96% of those claimants were not incurring much or any additional living costs (which is what PIP is supposed to be for). If people are being compensated for costs they don’t incur, and a particular type of claim is growing very fast, there is probably something not quite right. It is quite reasonable to try and tinker with the system to get it right.

The tinkering that the DWP decided to make, which led directly to the resignation of Iain Duncan Smith, was to reduce the score you get for needing an aid or appliance in the “going to the loo” and “getting dressed” categories. So grab rail for the bath would still be 2 points, but if you also need one for the loo (which you probably would), that’s just 1 extra point. This reduces the double-counting effect of two very similar measures. Anything you need for getting dressed also just gives you one point.

This change would only have affected people whose claims rely entirely on needing to use an aid or appliance to use the loo, get dressed and two (but not more than two) other activities. Anyone scoring points for other things (e.g. needing a carer to help), or in more than two other activities would still have scored enough points to receive PIP.

(Update 25/3/2016 in response to comment about dropping from enhanced to standard rate: Another consequence might be that someone currently scoring exactly 12 or 13 points, including 2 points each from dressing / toilet needs, would drop from enhanced to standard rate of PIP. Threshold for enhanced is 12 points. It is hard to know without the raw data how many people would be on exactly 12 or 13 points, but looking at the score cards it feels likely that someone racking up enough points to reach that threshold would have needs sufficiently high to score a lot higher than 12 or 13. They'd need be scoring something on one of the learning difficulty measures, in which case you'd imagine they'd score on other learning difficulty measures putting them well over 12 points. Or they'd need 2 points on all six of the other measures and nothing else ... but in that case I would still wonder whether their case merits the enhanced rate, actually. Regardless of all that, following para still applies: this sort of thing is hard to get right, and Gov't should be allowed to review their scoring system to try and get it right.)

The trouble with scoring systems like this is you need to test them out to see if you have got the weightings right. The independent reviewer suggested the weightings might not be right. The DWP looked into it and proposed a change. We should applaud a Government which actively tinkers with small details like this in an effort to target money at the right people. Instead we throw our toys out of the cot in a childish tantrum over “taking money from disabled people and giving it to the rich”.

And all the while we completely fucking ignore the Enron-style accounting tricks that our Chancellor deploys to contrive a surplus in 2019/20. Look at this graph (annotations are mine) from the Office of Budget Responsibility. It shows how the Chancellor manages to turn an expected deficit of £8bn into a surplus of £10bn:

(Source: OBR Economic and Fiscal Outlook March 2016)

That's what we should be shouting and screaming about.

What if we didn't bomb Syria? What if we'd never invaded Afghanistan or Iraq?

2015-12-02T14:08:00.000+00:00

Thought experiment: it is late September 2001. 9/11 happened a few weeks ago. USA is considering their response.

They decide to fix the inter-agency problems which weakened their security. They decide to massively increase intelligence and security spending. They create the Department of Homeland Security to ensure a single agency has over-arching responsibility for keeping their citizens safe. They mandate that airport screening staff are federal employees.

They keep a watch list of unsavoury characters they'd rather not have in their shores.

Their spies abroad monitor and infiltrate terrorist groups.

Other Western countries introduce similar measures.

But no-one invades Afghanistan and or Iraq.

The Taliban continue their unpleasant regime in Afghanistan, skirmishing with the Northern Alliance and others. Maybe some other warlord boots them out, maybe not. MSF continues to run hospitals in Afghanistan, and their doctors and nurses are not killed by US airstrikes.

Al-Qaeda continues to mount terrorist attacks against the West, but they lack widespread sympathy and their cells are often infiltrated and plots foiled. They operate just a little more "successfully" than those right-wing extremists preaching the overthrow of the American federal government: watched, sometimes taken seriously, and occasionally blowing things up, but no-one claims they are a 'threat to civilisation'.

Saddam Hussein continues his secular dictatorship of Iraq. He continues to imprison and torture political prisoners, but (if we are counting bodies) these number in the hundreds, or perhaps thousands. The hundreds of thousands of civilians killed during the US-led invasion don't die. Most people go about their daily lives in safety. Al Qaeda in Iraq never comes about, and nor does its successor, ISIS.

The West remains a beacon of human rights and freedom, as it was during the Cold War. The intelligence services do not fund or foment violent revolution against these unpleasant Middle Eastern regimes, so there is no Arab Spring. But the existing democratic structures in these countries slowly gain influence and the power of the dictators wanes, as they did in Eastern Europe.

(Syria has a democratic Parliament, remember? Their Speaker wrote to our Speaker a while back.)

Disenchanted Muslim youths still listen to radical preachers condemning the decadence of Western life and dream idly of converting the world to Islam. But there are no videos for them to watch of US helicopter crews shooting journalists, or of children maimed and killed in a Hellfire missile attack, or of Fallujah civilians apparently killed during a white phosphorus attack.

They don't watch these videos online, because these things don't happen.

So some are enraged and angry, but in the way that many young people are enraged and angry: enough to protest, but not enough to blow themselves up in the marketplace.

What happens when children from poor families get a little bit poorer still?

2015-11-02T18:12:00.000+00:00

There's good evidence about what happens to children from poor families if you give those families extra money. The children behave better and show more positive personality traits.

There is data on this from tax credit programmes around the world, but the best data comes from a long-term study in North Carolina. The study didn’t set out to look at the effect of giving money to poor families, but during the course of the study, some of the families received an unexpected windfall because they were Cherokee Indians who started to receive profit share from a casino built on their reservation. As a result, the researchers were able to compare those children who got unexpected extra money with those children who did not. The researchers found that the children in families who received the extra money showed fewer behavioural disorders and scored higher on those personality traits which are associated with better life outcomes.

Will we also, in future, find out what happened to children of poor families who, unexpectedly, lost a big chunk of their income after April 2016 because of the changes to tax credits?

Will there be a cohort of millions of children who behave worse and have poorer life chances because their familes found themselves with a lot less money after April 2016?

Time will tell.

There's a good summary in Washington Post of the Cherokee Indian research:

http://www.washingtonpost.com/news/wonkblog/wp/2015/10/08/the-remarkable-ways-a-little-money-can-change-a-childs-personality-for-life/

And here's the link to the original paper:

http://www.nber.org/papers/w21562.pdf

Need money for tax credits? Try putting corporation tax back to where it was.

2015-10-27T17:11:00.001+00:00

Corporation tax has been reduced from 28% to 20% (now) and then 18% (planned for 2020).

This would appear to be the fastest reduction in corporation tax since it was introduced in the 1960s (Souce: UK Corporation Tax Rates on Wikipedia). It will also give us nearly the lowest corporation tax rate in the OECD. The United States, for example, has a rate of 39%. (Source: OECD). I'd say this is a pretty radical policy.

The rationale behind this tax reduction is that cuts to corporation tax encourage companies to invest and encourage multi-national companies to put profits through the UK (if they can get a lower rate here than elsewhere). That's all well and good, but the fact is that if you cut corporation tax you get less tax.

A lot less tax.

I've added up all the OBR estimates for the impact of each reduction in corporation tax, and I'd say that, cumulatively, these add up to costing the Exchequer about £5-10bn per year. These estimates take into account the fact that the tax base will increase slightly as the tax rate is lowered (because multi-national companies choose to pay tax here rather than elsewhere). Click the image below to view my Google spreadsheet:

These particular OBR forecasts don't take into account any change in the overall economy (for example, if lower taxes lead to higher private investment and therefore higher economic growth). That's a bit harder to figure out, although there is plenty of literature on it, including a paper from HM Treasury which says:

(Source: Analysis of the dynamic effects of corporation tax reductions, HMT/HMRC)

Yep, you read that correctly. Slash corporation tax and you'll get a fabulous 0.65% uplift in GDP after 20 years, and you get about half of the cut back, eventually, through other taxes because of that extra economic activity.

But the Government still has less money to spend overall, and especially in the intervening years.

What if you look back to the OBR's forecast before these massive tax cuts were annnounced? Are we now seeing much lower corporation tax receipts than the OBR was originally expecting? Yep. Corporation tax receipts are about £10bn lower than the forecast from before the rate started to be lowered:

(Source: Pre Budget Forecast June 2010 & Economic and Fiscal Outlook July 2015)

Just before the 2010 Budget, OBR was forecasting about £51bn in 2014/15 from Corporation Tax ("Onshore" - i.e. excluding Oil & Gas). Now we are expecting just £41bn for 2014/15.

What could you do with that extra £10bn?

Spend it on tax credits for low-paid families, maybe?

It is misleading of the Government to claim that spending cuts are necessary because public spending is too high and unsustainable. Spending cuts are necessary because we have been cutting taxes.

Why the Budget Responsibility Charter is a dumb idea

2015-10-15T17:15:00.000+01:00

Plenty of economists will give you sound technical reasons why George Osborne's Budget Responsibility Charter is a bad idea.

But here's the simple version:

1. There is nothing wrong with Governments borrowing money to build big expensive things (like hospitals and railways) which last a long time. Sensible public works create economic growth and as long as the growth rate exceeds the interest rate on your debt, then it is a good thing to borrow money to do it. The alternative is to make taxpayers this year pay for something which might last 20 or 50 years. Or save up over 20-50 years and build the hospital then.

2. Government borrowing smooths out economic ups and downs. An example: let's say Government forecasts £700bn of tax revenue, but only gets £699bn because of some minor timing issue in tax receipts. What are you going to do? Borrow £1bn until the tax comes in? Or fire 30,000 nurses at £35k a pop because you stupidly promised never to borrow any money at all ever again?

I mean, FFS, it is just totally economically illiterate. As, hilariously, George Osborne himself said when the last Labour government suggested something similar back in 2010.

When Labour gov proposed Fiscal Responsibility Act, in Jan 2010, George Osborne was scathing. Take a look #newsnight https://t.co/2ke7on1Loj
— BBC Newsnight (@BBCNewsnight) October 14, 2015

Though I wish John McDonnell would articulate this himself more clearly instead of making vague assertions about wanting "growth not austerity". What he should be saying is: we think it is OK to borrow money for investment, and to smooth out the economic ups and downs. That was Gordon Brown's 'Golden Rule', which everyone seems to have forgotten about.

Has competition in energy supply actually worked?

2015-10-01T10:41:00.000+01:00

I am (used to be?) a free market entrepreneur. I should believe that privatised energy companies will provide a better price and service than a state-run monopoly. But, I have to admit it, the evidence seems to suggest otherwise.

Competition hasn’t made suppliers more efficient

If competition makes companies strive for lower costs, you’d expect all energy suppliers’ costs to be roughly the same (per account, or per £ of revenue, or whatever like-for-like comparison you might make). They should all gravitate towards the most efficient business model and enjoy similar costs to each other.

Yet Scottish & Southern, for example, incurs about £80 of cost per customer account, while EDF incurs almost twice that at £140. (Source: Ofgem, Figure 50.)

It is hard to explain this. It is not a question of scale (although if it was, that would be an argument in favour of the biggest-scale option of all: state monopoly).

And it is not as if the energy companies are still finding their feet and the strive for efficiency just hasn’t finished yet. This divergence in costs has been present for years and is getting worse, not better:

"the differential [between the most and least efficient companies] appeared to have widened, indicating that competitive pressure was failing to drive convergence in supplier costs, as should be expected according to economic theory." (Source: IPPR, page 26)

Whatever the explanation, this one detail makes it very hard to argue convincingly that competition is working in the energy market.

And if competition isn’t driving improved efficiencies, then there is nothing to counter the extra costs which privatisation creates such as marketing costs and paying returns to shareholders.

A monopoly has no marketing costs

Energy companies spend money marketing their services and paying price-comparison sites for leads. That marketing cost has to be built into the price the customer pays.

I haven’t found accurate figures for the amount of money that energy companies spend on marketing costs but they do pay about £50 for a lead from a price-comparison website (Source: Submissions to Energy & Climate Change Select Committee enquiry, para 19). About 10% switch every year, 10% * 25m households x £50 = £125m. Not much in the grand scheme, but a monopoly would not incur those costs at all.

A public monopoly has no shareholders to pay

The profits which energy companies pay to shareholders work out at approximately £50 per year per account. (Source: Ofgem, Figure 1). This is the money which the shareholders earn in return for providing the energy company with capital to run their business. There’s nothing inherently wrong with energy companies making profits. But the fact is that a state monopoly would not need to pay this return. (Well, strictly speaking, the monopoly would enjoy a much lower cost of capital as it is funded by the Government, who has a much lower cost of capital than any corporation, because Governments can access capital very cheaply.)

When energy supply was privatised, everyone knew of course that the privatised companies would incur higher marketing costs and higher cost of capital than a state-owned monopoly. The expectation was that competitive pressure should drive efficiencies which more than compensate for these higher marketing costs and higher cost of capital.

But that hasn’t happened, all we’ve got is the higher costs without (it seems) the corresponding drive for efficiency.

But public monopolies provide crap service!

This is the killer argument in favour of privatisation if all else fails: service from the government-owned utilities, back in the day, was crap, and we don't want to return to that.

But service from all companies in the UK in the 1970s was crap, remember? That's a 1970s thing, not a government-owned monopoly thing.

How, in practice, could the Government re-nationalise energy supply?

Jeremy Corbyn has suggested that a Labour government would re-nationalise energy supply (FT in August). Although it seems they've now dropped that idea (FT in September).

But would it be feasible to bring energy back into public ownership? Energy supply is not like the railways. With railways the government has granted time-limited franchises which the government can just take back when they expire.

The Government cannot simply declare a monopoly and prevent existing energy companies from operating. Well, they could, but that would be unfair on the shareholders of those companies, and would permanently damage the principle of private property ownership.

If an existing private company wanted to corner the energy supply market they’d have to acquire the energy supply business of each of their competitors. This kind of thing does happen from time to time: e.g NPower sold 770,000 accounts to Utility Warehouse in 2013 for £218m (Source: FT). That's about £280 per account, or 5x EBIT per account (industry average EBIT per account is roughly £50, see above).

There are approximately 25million households in the UK, so by the back-of-the-envelope calculation the total cost of renationalisation might be roughly £7bn (£280 x 20m) (about 0.5% of GDP, or about 7% of annual NHS spend in England or about 7% of the current annual Government deficit).

Could be twice that, but can't see how it can be 10x that.

The IMF says we should borrow money to fund public infrastructure investment

2015-09-17T13:49:00.001+01:00

A few days ago I put together this Downfall parody showing George Osborne (played by Hitler) learning that his austerity drive has slowed the recovery. He realises that the recovery might have been faster if he had maintained government spending levels.

One Twitter user made this comment after seeing the video:

@brucegreig @kevverage Perhaps someone will do one where you realise that growing the economy is not much use if you borrow to do it.
— voiceofpeason (@voiceofpeason) September 15, 2015

My reaction to this? Of course it is OK to borrow money to grow your economy. That's what debt is for: to fund investment so you can make more things, more efficiently. With interest rates so low, you only need a tiny improvement in the economy to pay for the interest on the debt. Seemed like a no-brainer.

I throw out this statistic (which I'd heard on a macroeconomics podcast while sprinting up St Catherine's Hill trying in vain to regain my PB):

.@voiceofpeason @kevverage e.g. Standard & Poor calculates that every £1 spent leads to £2.50 extra GDP http://t.co/xkvhHyUcRI
— Bruce Greig (@brucegreig) September 16, 2015

@brucegreig For this to work GDP needs to increase enough to produce tax revenue to lower debt ratio. But £1 debt = only £1 tax revenue.
— voiceofpeason (@voiceofpeason) September 16, 2015

@voiceofpeason Think not. *GDP* (not tax revenue) has to increase by more than the debt increase for Debt:GDP ratio to improve.
— Bruce Greig (@brucegreig) September 16, 2015

As this continued, I became less sure of my ground. I couldn't find this 2.5 multiplier widely quoted anywhere else, for example. So read up on it. And my reading took me to this excellent and very accessible paper from the IMF

The IMF is not known to be a bastion of crazy left-wing ideas so I felt confident I wasn't being hoodwinked into thinking something was mainstream when it isn't. And the findings are based on actual data from different governments' investment spending, rather just being some economist's ideas about how things might be working.

This paper concludes that an increase in government infrastructure spending of 1% of GDP increases GDP by about 1.5%, on average, after four years (figure 3.5 in the paper).

That's not as high as the 2.5x I'd quoted on Twitter, but this effect, according the to IMF, varies a lot depending on the circumstances. It is highest when the spending is financed by debt (rather than just coming out of day-to-day tax revenue) or at times when economic growth is slow. Under either of those circumstances, the same 1% of GDP spending translates into about 3% increase in GDP. (Figure 3.6, panels 1 and 9)

But what about the money you borrow? Does the growth in GDP really compensate for the higher borrowing? That is @voiceofpeason's basic question.

Answer: yes. Investment spending improves your debt:GDP ratio, especially if the spending is financed by debt.

Might need to repeat that:

Investment spending improves your Debt:GDP ratio, especially if the spending is financed by debt.

How can that be? How can more debt improve your debt:GDP ratio? Because debt-financed investment has a greater effect on GDP than budget-neutral investment. The GDP grows further if you use borrowed money to invest. Sufficiently further to outweigh the increase in borrowing. So your debt:GDP ratio improves.

(The paper doesn't try and explain why debt-financed investment should be more effective at growing GDP than investment that comes out of regular tax income. My layman's guess would be that if you use ordinary tax revenue for infrastructure investment, rather than borrowed money, you are taking money away from other purposes and that slows the economy somewhat. So the benefits of the investing spending are offset by the fact you are taking money away from other areas of the economy. If you use borrowed money, you are not taking it away from other government spending areas.)

Given that the IMF say this, why aren't governments doing it? The authors do comment on that, and say this:

"The reason is that in practice, public investment decisions frequently are not guided by economic rationale. Unproductive projects are often pursued by politicians and line ministries when they should not be, and some productive projects (and importantly, maintenance) are forgone when they should be given priority."

Translation: politicians, even those in charge of the economy, don't make decisions based on economics. That's a damn shame.

______________________

PS: This IMF paper is just talking about infrastructure investment. Roads, railways, bridges and so on. My Downfall parody talks about government spending generally, rather than simply infrastructure spending. The effects of different types of spending seem to be thought to have different multipliers (Wikipedia), but as long as the multiplier is greater than 1, then the effect of the spending will be to reduce the debt:GDP ratio.

For example:

GDP = £100bn

Debt = £100bn

Ratio = 100%

Multiplier = 1.01 (i.e just a tiny but more than 1) after 3 years

You borrow £1bn and spend it, taking your debt to to £101bn. Economy grows by £1.01bn, taking it to £101.01.

Ratio = 101 / 101.01 = 99.99% (i.e. lower than it was before)

Daniel Lacalle and the same old 'PQE would be inflationary' fallacy

2015-09-16T11:50:00.001+01:00

I'm cautiously supportive of Jeremy Corbyn. I signed up as a £3 Labour supporter to vote for him, for reasons outlined here.

And now I'm slowly trawling through his policy ideas to understand them better. This tweet from my former dotcom-era boss, Michael Liebreich, caught my eye:

Any student idealists who think @jeremycorbyn's #PeoplesQuantativeEasing might work should read this: http://t.co/nQEC7o8JhO
— Michael Liebreich (@MLiebreich) September 14, 2015

Michael is a pretty smart guy (Oxbridge, McKinsey, Harvard Baker Scholar, blah, blah) so I thought I'd take a look.

But the article Michael links to is disappointingly unsophisticated, even to someone, like me, whose macroeconomic education spans about the last two weeks.

The article, by a Spanish economist and author called Daniel Lacalle, repeats the same basic macroeconomic misconceptions that numerous mainstream commentators make. For example, Lacalle says the UK economy is doing really well now, vindicating the UK Gov't austerity policy:

But the 4.5% growth figure quoted is total growth since pre-crisis peak, not per year. 4.5% in, what, 6 years? That's why the Governor of the Bank of England described this as "the slowest recovery on record". On record. Ever. Maybe even since the South Sea Bubble of 1720.

When the Prime Minister tried to claim that his austerity policies didn't hamper GDP growth, the independent Office for Budget Responsibility wrote a stern letter to him setting him straight, confirming that austerity (aka "fiscal consolidation measures" in the letter) does of course slow the growth of the economy.

Even more crucial is the fact that growth per person has not recovered. Employment has grown, partly because of inward migration, so the total economy has grown. But there's no point growing the economy by creating more jobs if productivity does not improve. That's what the Luddites tried to do: create more jobs by destroying productive equipment.

If you look at the "recovery" in terms of GDP per capita, it looks truly dire on a 50-year view:

To me, the data on the UK economic recovery suggests we need a change of plan, not more of the same.

Next, Lacalle falls for the 'deficit spending caused the financial crisis' line:

WTF? Spain was running a surplus before the crisis hit. Whatever you think about the reasons for Spain's problems, you can't say it was because they were running a deficit:

Only after the global financial crisis hits does the Spanish Government position deteriorate as tax revenues fall and welfare spending rises. France was running a deficit before the crisis, but at around 3% of GDP was at the time considered perfectly acceptable and not profligate.

The extraordinary thing here is that Lacalle is Spanish. How can he not be aware that the Spanish government was running a surplus prior to the crisis? Maybe the dates he's used are a typo, and he means "deficit spending in the 2008 - 2014 period failed miserably" rather than the "2004 - 2010 period". If so, see earlier point about how reducing a deficit in a recession slows economic growth.

Incidentally, the UK Labour government, too, was spending very sensibly until the crisis hit:

Lacalle then moves on to look directly at People's Quantative Easing, the idea that the Bank of England should create new money for the government to spend:

Even the first sentence in the paragraph is misleading. The UK doesn't have a "policy of increasing the money supply aggressively". Private banks control, by making loans, the money supply (yes, banks create money, not governments, except in the special case of QE). The independent Bank of England then adjusts interest rates to manage inflation, that much is true. But they only target the headline inflation rate, they don't target house price or financial asset inflation specifically. And because most new money goes to mortgages or financial assets, massive increases in the money supply lead to asset price bubbles, even if the headline inflation rate looks OK.

So to claim that PQE is a bad thing because the current arrangements work well is shaky to say the least. Many people would argue that housing and financial asset bubbles are a bad thing and we should organise our economy to stop them happening.

Then Lacalle's actual criticisms of PQE fall pretty much to the level of the Daily Mail, namely that it would be inflationary and encourage the government to spend money on "white elephants":

Even as Corbyn outlines it, PQE would not provide unlimited money. Corbyn is not a simpleton, I'm sure he understands that unlimited new money creation by the state would lead to inflation. Just saying "PQE won't work because it will be like Venezuela" is not a helpful critique from an experienced economist and popular author.

PQE alongside the standard monetary framework (i.e. BoE controlling interests independent of government) probably wouldn't work, because in normal times any benefit of the newly created money would be offset by BoE adjusting interest rates to avoid inflation. As it stands, PQE wouldn't be inflationary, but it wouldn't make much difference to the economy either.

But it seems quite plausible the PQE will eventually morph into some kind of Sovereign Money proposal, where the Bank of England directly controls the money supply. Then, instead of having to react to changes in the money supply by changing interests rates, the BoE just decides directly how much money to create each month. Which is what the pressure group Positive Money have long been proposing, and they have a whole page of rebuttals to objections of a more sophisticated nature than Lacalle's.

Simple solution to Mediterranean boat carnage?

2015-09-14T15:35:00.001+01:00

Why are so many refugees spending thousands of dollars each on life-threatening boat trips across the Mediterranean?

Why don't they fly instead, which would be much cheaper and safer?

They can't fly because airlines would not let them board. Airlines have to bear the cost of returning illegal immigrants to their home countries. The onus is on the airline to check that someone travelling to the UK is entitled to come here. That'll be in Directive 2001/15/EC:

But, hang on a minute. What's that line at the bottom there:

An asylum seeker is not an illegal immigrant. Once they make a claim, they are legally entitled to stay while their claim is processed. They would not be refused entry, so the airline would not have to transport them back.

The directive allows airlines to transport asylum seekers. It doesn't say "granted international protection", it just says "seeking international protection". Merely claiming asylum is enough to indemnify the airline from the costs of returning the person.

Maybe this isn't watertight. Maybe the airlines would worry that the government would still come after them for the costs of deporting people whose asylum claims later fail.

So here's my plan: airlines could charge a very substantial insurance premium (say, €500) to people with plausible claims to asylum (e.g. anyone from Syria) and fly them here. Most people fleeing Syria will be granted asylum, so the airline will never be on the hook to fly them back. The airline can then pocket the extra €500.

Those whose asylum claims fail immediately and are turned away at the border would have to be sent back and maybe the airline would be sent the bill, but that's OK because they'd have a big pot of €500 fees which they charged everyone else.

This would, at a stroke, save thousands of lives.

And it wouldn't have to be an airline. Someone could do this independently: beat the people smugglers at their own game by chartering a 747 and flying it to Turkey.

Osborne's new Fiscal Rules are not economically credible

2015-09-03T20:00:00.000+01:00

The more I read up on macroeconomics, the more astonished I am about the things I learn.

For example, earlier this summer, the Chancellor published new rules designed to enforce prudent levels of government spending.

At first glance, they might seem quite sensible. They commit the government to eliminating the deficit by 2019/20. (For newbies: the 'deficit' is the shortfall between tax revenue coming in and government spending going out. The 'debt' is the accumulated historical debt built up over time.)

And then, once the government has eliminated the deficit, they will be required to maintain a budget surplus forever afterwards (i.e. spend no more each year than they receive in taxes).

Plenty of people (including me) might disagree with the overall thrust of this. Most economists would say that there's nothing wrong with governments borrowing money to, for example, build things which might last for a generation or more. Is George Osborne seriously proposing that are we really going to pay for a new generation of nuclear weapons, which will last for a generation, out of one year's taxes? Why not borrow the money and spread it over the useful life of the asset? That's what any sensible business person would do, especially if you control the world's fifth-largest economy and can borrow very cheaply.

But all that aside, even if you think a Government can survive in normal times without borrowing any money, there are some other worrying implications of these rules.

Governments plan their spending partly based on forecasts for how the economy will fare in this and future years.

So what if the forecasts aren't quite right? What if your forecasts show you'll make a small surplus, but something happens so you get a bit less tax than expected? Or have to spend a bit more on welfare than expected? Or have to spend a bit more on invading some Middle East country than expected?

The rules don't let you borrow money for that. So you'd either have to slash spending, or raise taxes.

(The rules only allow you to borrow money if the economy suffers a 'shock'. A shock is defined as GDP growth of 1% or less. Anything more than that and the rules hold.)

Let's look at the real implications of this, as regards eliminating the deficit by 2020. To eliminate the deficit, the economy needs to grow by 2.4% per year for each of the next five years (according to the Office for Budget Responsibility forecasts), and if it does that then the Government expects to have a surplus in 2019/20.

But what if the economy only grows by about 1.5% each year? Slow growth, but not a 'shock'.

That would leave the Government with a £20bn deficit in 2020 (Source: OBR via Jonathan Portes). The implication of the rules is that the Government would therefore be obliged to cut spending by a further £20bn, or raise taxes by a similar amount. Compare that to the pre-election fuss over finding another £12bn of welfare savings. It seems very unlikely that Government would actually do that.

Or, as Jonathan Portes puts it:

Are MPs like Zac Goldsmith and Peter Lilley and Michael Meacher in favour of People's QE?

2015-08-18T12:36:00.001+01:00

A lot of mainstream comment about Corbyn's People's QE has dismissed it as not financially credible. But it has cropped up in Parliamentary debates before, and several MPs and Lords would appear to already support it or something similar (i.e. some process whereby Bank of England creates money and spends it directly into the economy, sometimes referred to as 'helicopter money').

(Most of these comments arose in the "Money Creation and Society Debate" in November 2014, available on Hansard here)

Zac Goldsmith (Con)

"If the majority view is that quantitative easing is necessary, we need to ask this question: why not inject those funds into the real economy—into housing and energy projects of the kind that Opposition Members have spoken about—rather than using the mechanism in a way that clearly benefits only very few people within the world of financial and banking wizardry?"

Peter Lilley (Con)

"As a disciple of Milton Friedman, I am rather attracted to the idea of helicopter money; I think it was he who introduced the metaphor, and said that it would be just as effective if money were sprayed by a helicopter as if it were created by banks. "

Steve Baker (Con)

"Once the Bank legitimises the idea of money creation and giving it to people in order to get the economy going, the question then arises: if you are going to create it and give it away, why not give it to other people?"

Michael Meacher (Labour)

"The central bank would be exclusively responsible for creating as much new money as was necessary to support non-inflationary growth."

Austin Mitchell (Labour)

"If we go on creating more money through quantitative easing, we should channel it through a national investment bank into productive investment such as contracts for house building and new town generation."

Angus MacNeil (SNP)

"If QE goes into the pockets of those who are going to spend the money, surely QE can create some more motion in the economy, but if QE goes into already deep pockets and makes them larger and deeper, that is a very different thing."

Mark Durkan (SDLP)

"Quantitative easing has shown that if we are to use the facility of the state—in this situation, the state’s main tool is the Bank of England—to alter or prime the money supply in a particular way, we could choose a much better way of doing so than through quantitative easing. It is meant to have increased the money supply, but where have people felt that in terms of business credit, wages or the stimulus that consumer power can provide?"

And I found a couple of older references in the Lords from different debates:

Lord Skidelsky (Crossbencher)

"If we want to print money, I would direct it not to the banks but into the pockets of the British people. This is akin to Friedman's helicopter money. Here are two ideas: first, we could have a temporary wage subsidy paid to employers to encourage them to take on more workers; secondly, the Government might give every household a Christmas present in the form of a voucher to be spent on British goods within three months. There are many other methods, but the object would be to get the biggest bang of spending per buck of money."(Hansard)

Lord Newby (Lib Dem)

"To put it in layman’s language, he [Milton Friedman] argued in favour of providing “helicopter money”, whereby the Government print money and effectively drop it from the sky in order to get expenditure going."(Hansard)

Even the then Financial Secretary to the Treasury, Andrea Leadsom, doesn't dismiss People's QE out of hand, but presents a reasonable objection (that money created by People's QE cannot easily be later taken out of the economy, but money from normal QE can):

Andrea Leadsom (Con)

"[Normal QE can] at least it can be unwound, unlike the proposal for “helicopter money”, which would seem to be a giant step beyond QE—a step where money would be created by the state with no obvious way to rein it back if necessary."

Somewhat surprisingly, I'm voting for Jeremy Corbyn

2015-08-14T16:11:00.002+01:00

Somewhat to my surprise, I find that I agree with most (all?) of the policies which Jeremy Corbyn is proposing:

People's QE

Well, you read it here first, after all. Normal QE cost £375bn and only about 7% of that money, according to the Bank of England’s own report, ended up in the ‘normal’ economy (the rest just sloshes around in financial assets never reaching the rest of the economy).

If you spent that money directly on infrastructure projects like building hospitals, refurbishing schools, and what have you, then you would see 100% of the money in the real economy.

I haven’t seen a credible criticism of this suggestion. The casual riposte that it would be “inflationary” is very superficial. The amount of money created would still be decided by the independent Monetary Policy Committee, not by the Government of the day (unlike, say, in Zimbabwe).

At the moment, the MPC attempts to control the money supply indirectly by adjusting interest rates and encouraging banks to create more (or less) money. With people’s QE the MPC would simply be controlling the money supply directly.

Update 2nd September 2015: there are now plenty of critiques of this policy above just 'it would be inflationary': The main one being that if you do PQE all the time, the benefits would be immediately offset by Bank of England raising interest rates. The barebones description that Jeremy Corbyn has outlined wouldn't quite work, if private banks were still free to create money privately (as they do now) and the Bank of England was still trying to indirectly control the money supply by managing interest rates. But it would work very well if we moved to full-reserve banking and gave Bank of England direct control over the entire money supply. BoE would decide how much new money required in the economy and give it to the Government to spend. Very detailed point-by-point rebuttals of many common criticisms of this plan are here.

Trident

If you want the UK to be a strong and powerful nation for the sake of it, then you might want nuclear weapons. North Korea wants nuclear weapons for similar reasons: to demonstrate how important and powerful they are. In practice, we’d only ever need our independent nuclear deterrent if we fell out with both Russia and the USA at the same time. As long as we are allied to one or the other of them, we don’t need our own nukes. And if it started to look like USA and Russia might gang up on us, then how hard, really, would it be to build a few new nuclear warheads from scratch?

Less expensive housing

This is the elephant in the room for me. We have a system that, over the past 20 years or so, has made houses unaffordable for most people. In 1996 a first time buyer would spend 17% of their take home pay on their mortgage. By 2008 this was nearly 50% (and nearly 70% in London)

(Source: Bank of England Statistical Database and Nationwide House Price Survey, via Positive Money here).

In Winchester (where I live), a 3-bed ex-local authority semi costs over £300k. There's no way that someone on the salary of a teacher or policeman (c.£30k) could afford a house costing 10x their salary unless already own a house, or they inherit money from someone who owned a house.

Fixing this would require radical policies. Building more council homes might help. Abolishing fractional reserve banking would probably help too. Jeremy Corbyn isn’t actually proposing the abolition of fractional reserve banking, but it feels like the kind of policy that Jeremy Corbyn’s Labour party might go for.

The usual craptrap about offering "help to first time buyers" (i.e. using public money to further inflate house prices, or offering encouragement to banks to lend more money to inflate house prices) isn't going to make any meaningful difference.

Europe

Corbyn is the only candidate who might campaign for a No vote on Europe. The others have all ruled this out. I don’t think we need to be in Europe. We are the fifth-largest economy in the world: we can continue to trade with Europe without all the distractions and tensions that political union creates.

***

So I’m one of the several hundred thousand people who have paid £3 to sign up as a Labour supporter so that I can vote for Jeremy Corbyn as leader. How about that.

Austerity vs nominal GDP growth (graph)

2015-05-15T12:37:00.001+01:00

This graph from Yanis Varoufakis's blog is very interesting (Varoufakis is the Greek Finance Minister). It shows the correlation between structural deficit reduction, otherwise known as 'Austerity' (i.e. reducing government expenditure and / or raising taxes to balance the government books and reduce annual borrowing) and nominal GDP growth.

There appears to be an inverse correlation between austerity and GDP growth. More austerity = less GDP growth. In extreme cases, like Greece, massive deficit reduction is correlated with massive decline in GDP. And in countries who have not engaged in austerity (e.g. Estonia, according to this graph), GDP has grown strongly.

Correlation doesn't imply causation. It could be that poor GDP growth causes deficit reduction (although that seems implausible). Or something else is causing both these things. Or you might conclude that if a government takes a load of government spending out of the economy the economy gets smaller. That is, you might conclude that austerity suffocates an economy. And you would not be alone in thinking this.

Why is no-one talking about quantitative easing? £200bn of free money?

2015-05-05T21:41:00.000+01:00

Why are none of the political parties talking about QE at this election?

Between March 2009 and January 2010 we spent about £200bn (14% of GDP) on buying financial assets with newly-created money in the expectation of improving the economy.

According to the Bank of England's own report, this activity increased GDP by about ~~1%, or £15bn~~. 1.5%-2% or £20bn - 30bn. Yes, that's right, we created £200bn of new money and later found just ~~£15bn~~ £20-30bn of it sloshing around in the economy (source: The Bank of England Quarterly Bulletin Q3 2011)

Many sensible people, and some economists, might argue that it would have been better to give the money to the government of the day to spend however it saw fit. New hospitals. Schools. Roads. Wind farms. Nurses. Whatever, really.

If a government spends £200bn directly into the economy you can be pretty sure you'll find at least £200bn of new money in the actual economy. But instead we just gave it to the owners of financial assets to allow them to buy more financial assets.

Rather than squabbling over the minutiae of almost-identical tax and spending plans I wish some party took a more radical position on important stuff like this. £200bn is a lot of money: roughly one third of the entire Government spending in a year. Or about two years' worth of funding the entire NHS.

The most excellent pressure group, Positive Money, have much more on this issue here:

http://www.positivemoney.org/2014/06/waste-375-billion-failure-quantitative-easing-video/

(Positive Money quote £375bn as total for QE, and they are probably right, but my source is the Bank of England report linked to above. I presume the extra £175bn was created after that report had already said that QE didn't really work.)

(Corrected GDP % figure 22/9/2015)

The oft-repeated claim that Labour spent excessively and wrecked the economy isn't really very true.

2015-03-25T10:38:00.000+00:00

I'm not naturally a Labour-voting sort of fellow, but I get very cross when I hear that the last Labour Government supposedly wrecked the economy by spending too much. That just isn't really true, but is widely believed to be true.

This is the kind of thing I'm talking about:

We've come a long way since Labour's recession.Ahead of the Budget, SHARE this video to let friends know that Ed Balls and Ed Miliband wrecked our economy once - and would do the same again.
Posted by Conservatives on Tuesday, March 17, 2015

The final clips, where Ed Balls and Ed Miliband both say they don’t think Labour spending was excessive are presumably meant to make them look like fools. Everyone now "knows" that Labour spent far too much money, right?

But the data doesn't really support this claim:

Sources:

Taxes: Office for National Statistics, via UKPublicRevenue.co.uk

Spending: National Government Archives, via UKPublicSpending.co.uk

Furthermore, the Conservative Opposition at the time agreed with Labour's spending plans. Here’s what David Cameron said about Labour’s spending back in July 2008:

“[We will ensure] that, over time, the economy grows faster than the state, so spending falls as a share of national income and we can reduce borrowing and taxes on hard pressed families.After the splurge of recent years, the new spending totals for the years 2008-09 to 2010-11 show public spending growing by 2 per cent in real terms, below the 2.75 per cent trend growth rate of the economy.

This is why we are sticking to Labour’s spending totals.

Taken alone, these are tight.”

Yep, the Conservatives considered, in 2008, that Labour’s spending plans were ‘tight’. Not profligate or excessive. But 'tight'. Which is pretty much what they are ridiculing Ed Balls and Ed Miliband for saying now.

(That speech seems to no longer be on the Conservative party website, but the Wayback Machine has it stored forever here:

http://web.archive.org/web/20100326151219/http://www.conservatives.com/News/Speeches/2008/07/David_Cameron_Speech_to_the_CBI.aspx)

Learning with spaced repetition

2014-06-26T13:08:00.000+01:00

Earlier this year, I discovered a memory technique called Spaced Repetition while researching how to learn Moroccan Arabic as efficiently as possible. I came across a blogging polyglot called 'Benny The Irish Polyglot' who recommends using this spaced repetition technique (along with several other language hacking tips.)

My conclusion: spaced repetition is astonishingly effective.

I was able to learn enough Moroccan Arabic words (assisted by a few hours of Skype tuition from teh most excellent Jamal Idrissi on iTalki) to be able to converse with locals almost entirely in Arabic when I went windsurfing in Essaouira in May.

But until I started researching about language learning, I'd never heard of spaced repetition. And I have a degree in psychology (and there is plenty of academic literature on it.) So you can be fairly sure that the vast majority of people who need to learn things by rote probably haven't heard about it either.

And people do need to learn things by rote. Schoolchildren need to learn spellings and multiplication tables; candidates for the Indian Civil Service exam need to know vast numbers of facts about Indian history and politics; lots of professional exams require at least some rote learning (for terminology, abbreviations, formulae, map symbols and so on).

The go-to app for spaced repetition is an open-source product called Anki. That's what I use for my Moroccan Arabic. But it is quite time-consuming to set up, and of course you'll only find out about it if you know about spaced repetition in the first place.

So, I'm going to build a series of apps for specific audiences who need to learn stuff by rote (and who would probably be prepared to pay for tools to help them).

Starting with: people who are training to be London Taxi drivers, aka "Knowledge Boys". To learn The Knowledge you need two things: a mental map of every street in London, and the location of thousands of points of interest. Spaced repetition won't help with the mental map of London, but it can certainly help you remember what street the Wallace Collection is on. Or the Worshipful Company of Glaziers. Or the Horse Hospital.

There are a few thousand people every year who apply to become London taxi drivers and embark on learning the Knowledge. They'll find a handful of knowledge apps in the App Store which help with 'Pointing' (as it is called) but as far as I can tell those apps aren't using spaced repetition, just normal flashcards.

So, I'm building a better Pointing app, using spaced repetition, and you can read more about it here:

The Taxi Knowledge App, an iPhone app to help prospective London taxi drivers learn their Points.

Lego pendulum clock

2014-01-20T14:08:00.002+00:00

I've quite got into clocks recently, after bringing back to life an old non-working grandfather clock which I inherited.

Here's a clock which I've built in Lego:

Credit to Ben Van der Waal for the beautiful and elegant escapement design.

Here are the gear ratios:

Gear Ratio Multiple Time per revolution (s)
____________________________________________________________
Escape - - 8
1 40:8 5 40
2 40:8 5 200
3 24:8 3 600 (This is drive wheel)
4 24:8 3 1800
5 24:12 2 3600 (Minute hand)
6 24:12 2 7200
7 8:8 1 7200 (Needed to reverse direction)
8 16:8 2 14400
9 24:8 3 43200 (Hour hand)

Drive weight is 2.4kg. It feels like the Lego could hold a much heavier weight, which could then move even slower and so last longer. But haven't tried that yet. If I can move the drive wheel to gear 4, it will last 3 times longer, roughly 12 hours (it lasts about 4 hours now).

Getting to 8 days (the duration of a typical grandfather clock) feels completely unachievable, but maybe with incremental improvements to reduce friction it is possible in Lego, we'll see. Or perhaps it would be possible to have several weights with a new weight engaging once the first has reached the ground.

Telegraph journalists in England have lower levels of basic skills than their grandparents

2013-10-09T13:53:00.000+01:00

Right. This has really annoyed me.

Lots of papers today reporting result of latest OECD skills survey, and all claiming that school leavers have lower basic skills than their grandparents. For example:

That just seemed to me kinda unlikely. Schooling was fairly rudimentary for many people 50 odd years ago.

So I checked the stats, which are available here: http://piaacdataexplorer.oecd.org/ide/idepiaac

Sure enough, in both literacy and numeracy, 16-24 yr olds in England performed at pretty much the same level as 55+ year olds:

Then I went back to the Telegraph article to check what numbers they were using. Pretty much the same numbers, in a nice little graph, showing that 16-24 year olds performed at, yep, roughly the same level at 55+ year olds:

So the graph in their own article clearly refutes the assertion made in the sub-headline.

Two more things:

1. One reason why other countries are seeing their young people out-perform their oldies, is because their oldies aren't as smart as our oldies. E.g. German 55+ average literacy score is 254, vs 265 for England.

So it is harder for our youth to outperform our oldies because our oldies are a lot smarter than, say, German oldies.

2. On problem-solving (not mentioned in Telegraph article), our young people kick butt compared to the older generation:

Just sayin'

VoxSciences turns your voicemails into emails

2012-06-14T08:29:00.000+01:00

I've been looking for a while for a way to get Skype voicemails sent to me by email. Listening to voicemails is time consuming and fiddly. Skype's iPhone client for some reason often fails to play voicemails, so I have to login from a PC to get a voicemail. Then I have to actually listen to it to find out what it is about and how urgent / important it is.

I tried SpinVox via Skype, but I found that was unreliable (three voicemails in one day were totally ignored by SpinVox, no notification and no transcription. I only found them by logging in to Skype from a PC). And SpinVox only sends them as a text, not an email. Voicemails are usually important, but are rarely so urgent they need to be texted.

I found another service called VoxSciences, and it seems to work really well. It does two things which SpinVox doesn't:

1. Sends transcript by email
2. Attaches an MP3 of the voicemail to an email (so if the transcription is uncertain, I can easily listen to the original recording myself)

(Works with any telephone / VOIP, just forward unanswered calls to your VoxSciences voicemailbox and caller can leave a message which will be transcribed.)

Costs £5 for every 30 messages (with minimum fee of £5/ month).

I've no relationship with VoxSciences, but I'd highly recommend them to anyone who wants their voicemails transcribed and emailed to them.

www.voxsci.com

Printing numbered raffle tickets is harder than it sounds

2012-02-21T11:24:00.000+00:00

I spent the best part of an afternoon last week battling with Microsoft Office to print 500 raffle tickets for our local playgroup fundraising.

You'd think it's easy, right? Create one ticket in Word, do a simple mail merge to pull in 500 numbers, then merge to printer.

But it is really hard (in fact, I think it is not possible) to get Word to print several differently-numbered tickets on one sheet of paper.

Anyway, to save anyone else that pain, I wrote a whole blog post explaining how I eventually did it. And I even set up a whole new blog, just for that post, to make it as easy as possible for fellow raffle-ticket-printers to find it:

Instructions for creating and printing numbered raffle tickets on your own printer.

Send invoices via InvoiceHub so your clients don't have to rekey your invoice into their accounts software

2012-01-17T10:25:00.000+00:00

I have a new project called InvoiceHub.

InvoiceHub accepts invoices, via email, from anyone. It extracts key data (invoice number, amount, etc.) from the invoice. Then it forwards the original to its intended recipient with a link inserted in the email to allow the recipient to view the extracted data, and zap it straight into their accounts software.

Here's an example:

Let's say I'm creating an invoice to send to a client. It doesn't much matter what software I use to create the invoice, as long as that software can produce a PDF which I can email to my client. Here's me creating an invoice in Xero. I'm invoicing a client called Eskimo Joe, for some supplies of ice:

I email that invoice, from Xero, to my client, Eskimo Joe:

His email address is eskimojoe@gmail.com. But rather than send it direct to his email address, I send it via InvoiceHub, to eskimojoe+gmail.com@invoice-hub.com.

When Eskimo Joe checks his email, he gets an email that looks pretty much as if I sent it direct to him. It has my invoice, and my text, and comes from me:

But, actually, it's already been through the InvoiceHub system, and InvoiceHub has already extracted the data from the invoice, ready for Eskimo Joe to import into his accounts software. Look at the link at the top of the email:

That link has been inserted into the email by InvoiceHub. The beauty of it is, though, that if Eskimo Joe doesn't care much for automagically importing stuff into his accounts software, he can just ignore that link and open the regular PDF attachment. But if Eskimo Joe clicks the link, this is what he'll see:

My invoice on the left, with key data (invoice number, date and so on) already extracted. All Eskimo Joe has to do is give InvoiceHub permission to import this invoice into his accounts system. Let's say Eskimo Joe uses KashFlow. He pops his KashFlow username and password on the screen here and InvoiceHub fetches his list of suppliers and account codes, so Eskimo Joe can say where he wants this invoice to go:

Then he clicks "Import to KashFlow". That's all he needs to do. When he later looks in his KashFlow account, the invoice will be already there:

Neither I nor my client needed to set up anything in advance. All I did was send the invoice to eskimojoe+gmail.com@invoice-hub.com instead of eskimojoe@gmail.com. Try it yourself, and let me know what you think. Is this a useful tool? Or just a solution looking for a problem?

Why are XML invoices not commonplace?

2011-11-17T10:00:00.000+00:00

I'm quite intrigued by the idea of being able to send and receive invoices electronically. Big corporates have been exchanging purchase orders and invoices electronically for decades, but almost no SMEs do this (except for those who are required to, in order to bill their big corporate clients. But often that doesn't work out so well.)

On the face of it, it should be quite simple to include an XML version of your invoice in some standard format which any accounting software can read. Pretty much everyone is using accounting software to create invoices, but they then flatten that data by printing or emailing a PDF, only for the recipient to have to painstakingly type it all back into their software. Would be better if they could just have the raw data go straight into their accounts software, without any need to retype it.

I stumbled across one attempt to create a standard for this kind of invoice data exchange, called SimpleUBL. "UBL" is "universal business language", which looks like a horrendously complex standard for exchanging all sorts of business process information. SimpleUBL is trying to make that simpler. But it still ain't simple as in iPhone. Browsing through it, I'm seeing hundreds of "Common Aggregate Components", things like:

PlacardEndorsementType

OwnerTypeCodeType

LongitudeMinutesMeasureType

Does that stuff really need to be in a standard used by SMEs for exchanging invoices?

And the whole point of XML is that it is supposed to be human-readable, but even something as simple as displaying amount before tax and amount after tax is a little cumbersome in SimpleUBL:

<cac:LegalTotal>
<cbc:LineExtensionTotalAmount amountCurrencyCodeListVersionID="0.3"amountCurrencyID="USD">479.50</cbc:LineExtensionTotalAmount>
<cbc:TaxExclusiveTotalAmount amountCurrencyCodeListVersionID="0.3"amountCurrencyID="USD">479.50</cbc:TaxExclusiveTotalAmount>
<cbc:TaxInclusiveTotalAmount amountCurrencyCodeListVersionID="0.3"amountCurrencyID="USD">527.45</cbc:TaxInclusiveTotalAmount>
</cac:LegalTotal>

Seems to me that it could be a lot simpler still, and still be useful for the majority of businesses, and be much easier to implement.

If you are actually typing details from a supplier's invoice into your accounts software, the key things you type in are the invoice number, the date, and the amount. If you are VAT registered, then you’ll also be splitting out the total before and total after VAT. So the absolute minimum you need would be something like this:

<Invoice> <InvoiceNumber>10011</InvoiceNumber> <InvoiceDate>2011-10-12</InvoiceDate> <AmountBeforeTax>100.00</AmountBeforeTax> <Tax>20.00</Tax> <AmountAfterTax>120.00</AmountAfterTax> </Invoice>

But you also need to record who the invoice is from and what it’s for. That’s a little trickier for the person creating the invoice to include usefully in their XML, because they don't know how they are represented in your accounts system, and they have no idea what your chart of accounts looks like. You might put a phone bill under “Office Overhead”, “Phone, Internet, Power”, or “Rent etc.”

Nevertheless, that information has to appear in the invoice, so your bare-bones XML invoice might start to look like this:

<Invoice>
<From>British Telecom plc</From>
<To>Fictitious Startup Ltd</To>
<Description>Telephone service</Description>
<InvoiceNumber>10011</InvoiceNumber>
<InvoiceDate>2011-10-12</InvoiceDate>
<AmountBeforeTax>100.00</AmountBeforeTax>
<Tax>20.00</Tax>
<AmountAfterTax>120.00</AmountAfterTax>
</Invoice>

I think in the UK you have to show the VAT % as well as the VAT amount, and there are probably other bits and bobs I haven't thought of, particularly where there are several different items on one invoice.

But my point is it just doesn't seem that hard.

Seems a little surprising that we are still having to email PDFs to each other rather than sending raw data. Is there a business here? Some kind of hub which accepts an invoice in any format, converts it to XML and zaps it over to the recipient's accounts software? Maybe.