Public services in a digital future

Being honest about the choices we face

This blog is part of a collaborative Joseph Rowntree Foundation project examining and exploring ideas we need to advance Social Justice in a Digital Age.

“Productivity isn’t everything” wrote Paul Krugman, Nobel prize-winning economist in 1990. “But in the long run”, he went on, “it’s almost everything.” Getting more output for each unit of input sounds great. That’s why lifting productivity growth has been a central ambition of every government in the post-war era. More productive economies enable higher living standards, not only in those sectors or firms delivering it, but for everyone.

Some sectors are harder to make more productive than others. To attract and retain workers in these less productive sectors from leaving to join more productive ones, something close to the real wage growth seen in the broader economy needs to be offered to them. In other words, workers in less productive sectors will enjoy real wage gains despite productivity growth being so hard to achieve in their sectors versus others. This was Baumol’s central insight. In this way Baumol might be regarded as the godfather of trickle-down – or at least trickle-across – economics. But, as James Plunkett explains, Baumol either missed or underplayed the ability of institutions and of government to get in the way of his ‘iron law’ holding true.

In the United Kingdom, high-touch human sectors for which productivity gains have been most elusive – social care, childcare, healthcare, education – are concentrated in areas dominated by public provision. This is the result of a political choice: as a society we have decided that there should be some universal floor for human dignity. And while more resource-intense versions of these public services are available through private channels, the universal floor is one that could not be afforded by all if not provided through the public sector.

When we look to the Office of National Statistics’ estimates of productivity growth in the public sector, the results are depressing. Outputs have lagged the rising input costs across the aggregate public sector over the past twenty years, and particularly so in childrens’ social care, adult social care, and education. Is this the dead hand of the state in action? Might putting public services entirely into the private realm deliver productivity improvements that are so desperately needed? Data from the United States does not suggest that this is the case.

Figs 1-2. Labour productivity and hourly compensation in the US nonfarm business sector, and hospital subsector, 1995-2021

Source: Bureau of Labor Statistics, October 2022

The US healthcare system is a hideously complex affair. Treatment is provided by the private sector, with different blends of private insurance and public programmes picking up the tab. It is a system in which strong commercial incentives to innovate and deliver quality care at lower prices are overwhelming, with multi-billion-dollar prizes on offer to those best able to build a better mousetrap. And yet in the highly commercialised and competitive hospital sector, productivity growth has been essentially zero since 1995 (figure 1). The wider private sector has, meanwhile, seen transformative productivity growth. Despite this vast differential in productivity outcomes, hourly compensation in the hospital sector has moved in concert with the wider private economy (figure 2) – just as Baumol would’ve predicted.

Furthermore, we can look to private sector doppelgängers in the United Kingdom like the private school system to see how much value is ascribed to a high touch human sector by those who have means. 

Over recent decades the proportion of children attending private school has remained relatively stable at c7-9%. And the ratio of the top 7-9% versus median post-tax income incomes has fallen rather than risen over the last twenty years. But spending per pupil on private day schools has increased at a far quicker pace than it has in the state sector, with the ratio of spending per pupil in the private sector versus state sector rising despite large real terms increases in state sector spending per pupil.

Fig 3. Comparing the Average Spending per Pupil in English Secondary State Schools to UK Private Day Schools in Constant 2020-21 Money, 1978-2021

Source: Institute for Fiscal Studies, Independent Schools Council, Office for National Statistics, author’s calculations.

Spending per pupil in the State versus private educational realm might provide a sense as to the degree to which our aggregate ability to afford better public services is being suppressed by government’s desire to constrain spending. Other complicating factors may be in play – for example, the degree to which private education consists of its signalling value (like a Prada handbag), or the degree to which the value of elite education is positional (helping to win the CV arms race). But despite these complicating factors we can see that the price paid for secondary education varies meaningfully depending on its limiting factor: the ability and willingness to pay the direct financial costs for private schools of one’s own children versus the collective ability and willingness – mediated by government – to pay for state schooling of other peoples’ children. And where government mediation is absent, an increasing share of income is spent – just as Baumol forecast.

What are the choices faced by society in thinking about how to manage high-touch low-productivity sectors dominated by public provision. There are three, and only three, ways forward – or some blend thereof. These three ways can be thought of as three points of a triangle of possibilities (figure 4). Let’s follow James in making this specific about a sector – social care.

Fig 4. The Triangle of Possibilities

The first choice is to pay more for social care. The proportion of national income going to social care would – taking this path – increase, and each social care worker would be paid more for each hour they worked (as the monetary value of social care increased). This is what Baumol would have predicted. What does this mean in practice? It means higher taxes, a larger state, and it could also open the door to better care. The productivity gains that are delivered in other parts of the economy allow for an overall improvement in our collective standard of living, and that universal floor for human dignity can rise. Realising such ambitions is arguably the point of economic growth.

The second choice is to get less social care – either in quantity or quality. This is what would happen if the buyer of social care is unwilling or unable to meet the rising costs and sets of a reverse auction for care (eg, takes as much social care as can be bought for a fixed amount of money; this might be less and less each year). When ministers stand up and talk about holding budgets flat in real terms, we might be tempted to think that this holds the level of care flat in real terms, but the Baumol effect means that a real terms budget freeze translates into a creeping reduction in services. On the ground this might manifest as shorter and more hurried care visits, with more clients assigned to each carer. It might manifest in the juniorisation of care work – passing work that has required significant training to the untrained or less-trained. It is even possible that such developments are recorded productivity improvements if the tricky judgements required for hedonic calculations mismeasure any juniorisation.

The third choice – although it’s not really a choice – is to work smarter and get more care for the same or less money. It is to build the better mousetrap that has eluded those seeking billion-dollar prizes for doing so in places like the United States with different models of provision. There’s something to be said for this. We shouldn’t get fatalistic about the absence of productivity growth in high-human-touch sectors. Just because these sectors have not yielded great advances before, it doesn’t mean they won’t in the future. Concerted human ingenuity has yielded incredible developments, and the door is open to innovation in the field of public service provision. But we must also prepare ourselves for the choices that arise from a continuation of the trend of the past: flatlining productivity and Baumol’s insights around the requirements to pay more for the same thing, or pay the same and get less in a growing economy.

There is no escape from the triangle. And we currently talk as if there is. So we really need to be more honest.

What story does the data tell about choices that we have made so far for social care in the UK? While never a highly paid sector, the median worker in UK Social Care in the late 1990s was paid a comparable hourly rate to the median non-social care worker. Two decades later this median social care worker earns only around three-quarters of the median worker outside social care (dark blue line, fig 1). This looks to be a direct result of the desire on the part of the government to – on the one hand – guarantee a universal floor for human dignity, but also limit the overall cost to the public purse. As such, the kind of dynamics one would expect from Baumol that we’ve seen in the US hospital sector – bereft of productivity gains – can be seen to have been constrained by the desire of the government to escape the triangle (figures 5-6). But there is no escape. Reducing the relative wage has manifested in juniorisation, recruitment difficulties, and an exodus of care workers into the retail and hospitality sector.

Figs 5-6: Ratio of hourly gross pay for UK Social Workers without Accommodation versus All UK Workers (Medians and Means); Hourly wages for workers in the US hospital sector, US private sector, UK social work sector, UK all employees 1997-2021 (1997=100)

Source: Bureau of Labor Statistics, Office for National Statistics, Authors Calculations October 2022

The bottom right corner of the triangle – higher productivity – which is so compelling because it avoids some really dramatic and unpleasant choices, looks so far to be based in fantasy. An hour of childcare/ eldercare today is comparable to an hour in 1870.

And so the big choices put us back at the opposite edge of the triangle. On this edge of the triangle there is a straight trade-off between a larger state that harnesses our aggregate economic productivity gains to maintain or lift the universal floor for human dignity on the one hand, and sequential reductions in quality and quality on the other. Whichever choice we make should be made in the open.

What would it really mean to succeed here and occupy the high productivity corner? Dystopian fantasies in which children and elders are plugged into VR headsets where they can be entertained, educated and stimulated in the metaverse, while robots operate in the physical realm to support their bodily functions or restrain them can surely be discounted. Increasing the number of children a childminder can legally look after each year would only deliver a deterioration in quality masquerading as higher productivity (ie, the bottom left corner). It would require a more granular view of things like work organisation, job design, market structure, business models; all considered for different geographies and sub-markets.

More efficiency is – and should be – a real goal, but retaining the centrality of human touch in these sectors poses limits to productivity gains that can be achieved.


Liability Driven Investment (LDI*) – quick explainer

I wrote a quick blog for FT Alphaville earlier about some of the frankly incredible moves that have been happening at the long end of the yield curve earlier today. I’m worried that it’s a bit too impenetrable to non-bond-geeks. So this is my attempt at a public service explainer. It’s still pretty bond-geeky, so you’re not being stupid if you don’t follow. You’re being human.

Imagine I’m a scheme with £100 of assets and £100 of liabilities.

I put £20 into ‘matching assets’ – say, long gilts.

I put the remaining £80 into low-volatility growth assets that might beat cash like short-dated bonds, private credit, diversified growth funds and equities.

The value of my liabilities are estimated by discounting them back to ‘present value’ using long-dated bond yields. Given my asset allocation of 80:20, I have a big asset-liability mismatch. Every time long-dated bond yields fall, the present value of my liabilities rise – but only a portion of my assets rise in value. While my assets are not affected hugely by changes in bond prices/ yields, my liability value is: I have a big mismatch that leaves my pension deficit subject to the whims of the bond market.

To hedge against this risk that bond yields move, I enter into an interest rate swap, receiving fixed for 20yrs and paying floating on £80.

Now my assets and liabilities are duration-matched. I no longer care whether bond yields go up or down: my funding ratio is hedged. If bond yields go down a lot it will inflate the present value of my liabilities and also my matching assets (the £20 of gilts) and my interest rate swap overlay (the £80 notional).

What security does my swap counterparty have against me going bust and not being good for paying floating/ receiving fixed over the next 20 years? I need to post some high quality collateral. Luckily I have £20 in long-dated gilts! This is more than enough, and more than I am ever likely to need. Most of this will count as ‘excess collateral’.

If yields rise, this eats into my ‘excess collateral’ (as more collateral is required by my counterparty). I might need to sell down some of my growth assets to replenish it.

This happens all the time. It’s no big deal. Sometimes rebalancing favours purchases of growth assets, sometimes it favours sales.

Rebalancing might happen daily, but more often it’s a monthly or quarterly affair.

What happens if there is a very sharp move in yields intra-month? Conservative risk management on the part of LDI-managers ensures that schemes are protected against these moves by conservative over-collateralisation, the levels of which are derived from rigorous stress-tests.

What happens if the move is so huge that your stress-tests are busted? You probably need to get scheme sponsors to sign-off with wet ink that you can do large asset sales in other areas (maybe liquidating assets held by other managers), or get them to stump up new cash.

What happens if you can’t get the logistics to work (maybe because assets are tied up in private credit, other illiquid stuff, maybe because you just can’t chase down signatories)?

You, the pension scheme, will have defaulted. Under the terms of your ISDA, your counterparty is likely to liquidate your collateral and close out your 20yr swap position. (I honestly don’t know what happens legally after this, but defaulting on financial contracts is not great.)

Where does this leave you? You are now unhedged against rate moves.

This may be great for your funding ratio (if yields soar, albeit this could trigger waves of others to be cleared out of their swap positions).

This may be a funding rate disaster if yields collapse.

I feel very very sorry for any pension fund that has just been whipsawed – losing their hedge at the intraday yield peak only to find yields 100bps lower. This would be their true disaster scenario.

Lastly, I want to give a HUGE thank you to the wonderful and brilliant Dame Kate Barker. It was actually her who set me off looking at LDI and potential systemic risks attached back in June. Always listen to Kate!!

* This is not about what John Ralfe would call LDI.  It’s with what John Ralfe would call ‘levered LDI’. Universally otherwise known as LDI.

Voter suppression in the UK?

The Guardian scoop today on photo ID requirement in UK elections references a @commonslibrary report estimates that 3.5m UK adults lack photo ID required to vote.

In the US, the Republican Party has been accused of playing this move to exclude Black voters from voting, and the accusation looks pretty compelling.

My gut reaction, like those trending across Twitter, was that the move to require photo IDs to vote in the UK is the Conservative Party’s version of voter-suppression.

That gut reaction might well be right, but it doesn’t look like it is supported by ONS Census data.

There appear to be three dimensions of exclusion/ suppression that are trending: age, ethnicity, and wealth.

First, age. It’s not immediately clear to me whether the mechanism for youth voter-suppression is thought to be the additional friction involved in the voting process, or because it is assumed the young have lower access to photo ID.

Young people are not very good at voting today, so I can see that adding additional hurdles might be reasonably likely to worsen youth turnout further.

Second, ethnicity. This tweet from @jasebyjason was matched by many other good tweeps

The stat looks to be sourced from this @electoralreform story saying 48% of Black people don’t hold a full driving license.

Third, wealth. Passports in the UK are expensive things, and it seems reasonable to guess that passport ownership might be proportionate to wealth.

There’s also the direct partisan skew about which @jburnmurdoch and @chrishanretty have done research (which is quite striking!!)

What can we say about the evidence behind this gut reactions from some quick Googling and pivot-tabling?

Well, the @ONS does a Census every ten years, the last for which data is available back in 2011. The 2011 Census includes a question about holding a passport – separate from nationality. I looked to see whether there was immediate data that would answer how ownership split by 1) age; 2) ethnicity, and 3) wealth, expecting that voter suppression of the young, non-white and poor to show up.

Someone put in a request for answers to this to be broken down regionally by age, sex and ethnicity back in 2016. Wealth data isn’t available, although someone smart with time on their hands could probably do something clever with regional incomes and infer it from the regional split. The data is found here.

The data did not match my priors.

I might’ve made a mistake somewhere, so please correct, but the following things can be drawn from the data:

  • Of c.65.5m people in the UK, 74% are adults with British, Irish or Commonwealth nationalities who I infer are eligible to vote in UK elections. The remaining 26% splits into minors (22%) and non-eligible adults (4%).
  • Of these 74% eligible voters, a full 15% don’t have passports (ie, 9% of the population are adults without passports, 64% are adults with passports). This is not to say that they might not have driving licenses or other photo ID.

How does the age break-down of passport holdings break down? It looks like the old are the ones who are under-represented. Close to 20% of those aged 60 or over lacked a passport compared to less than 10% of 18-40yr olds:

How did the ethnicity of those lacking passports break down? It looks from the data like those self-identifying as white are under-represented as passport holders.

How about the intersection of age and ethnicity? It looks like younger non-white adults have a higher likelihood of not having a passport than older non-white adults. You can’t really tell from the chart, but just under 4.5% of non-white adults under the age of 30 lacked a passport, while the proportion according to the ONS is around 2.5% for the over 50s.

What does this all mean?

  • Does this make requiring photo IDs good? No.
  • Does this mean that requiring photo IDs will not deliver a partisan skew to voting? No.
  • Does it mean that the proposal isn’t a cynical attempt to gerrymander demographically? No.

All this means is that if someone designed a requirement to have photo ID to vote based on the understanding that it would exclude younger and less white adults, this understanding does not appear to be aligned with the ONS Census passport data question.

Or the data is wrong.

Bitcoin’s energy debate (and the Nature of Money)

Joe Weisenthal‘s morning notes are always worth a read. This morning he touched on a topic that I’ve been thinking on for a few days – the arguments around Bitcoin’s electricity usage.

The electricity consumed by the Bitcoin network is a bit more than Pakistan’s – a country of >200m people. Is this worth it? I went down this rabbit hole after tweeting this out a few weeks back:

You can browse the replies at your leisure but you can see that there was not a lot of love from the Crypto Bros. I tend to use Twitter as a learning resource, and in between the abuse I got some nice links to counter-arguments. They took me back to my old favourite subject: the nature of money.

Money is the medium that makes otherwise incongruent things comparable from a market perspective.

We all attach our own and different subjective values to different marketable things, but a marketable thing’s monetary value represents the crystalized triangulation of all of our subjective values (weighted by our individual monetary means). Simmel wrote a lot about this, basically outlining money’s role as reifying societal value.

So while the subjective value/ utility that I ascribe personally to a Bitcoin might be lower than that to which I ascribe a glass of clean drinking water, the money-weighted triangulation of subjective values/ utility ascribed to a Bitcoin is much much higher at around $46k.

Mark Carney outlined in his recent Reith Lectures that this money-centred value theory is actually somewhat new. In the late nineteenth century and early twentieth century, a group of economists known as neo-classicists launched an upheaval in value theory – comparable to the Copernican revolution in science. For older theoreticians like Smith, Ricardo and Marx, markets determine the distribution of value which was derived from labour. According to the neo-classicists, labour doesn’t give a good its value; instead, labour is valued because it is an input to a good onto which people project value. We can see how much value they project onto a good or service in its price (which serves as a money-weighted triangulation of subjective value). As such, price became the best description of value. Or, too take this to the extreme, Price = Value.

Going down this line of argument, the resources are consumed in delivering Bitcoin Proof of Work have been bid away from other projects because PoW has a higher value attached to it than any other activity. And as such it is idiocy to focus on electricity used by that which the market has determined to be the highest value activity.

Another line of defence in the energy use of Bitcoiners is that electricity isn’t fungible, and so Bitcoin mininng effectively utilises stranded or unreliable generation assets and turns them into something considered useful by many rich people (ie, Bitcoin). Some advocates sometimes go on to say that Bitcoin mining arbitrages electricity costs, although I think that this pushes the argument too far because energy that is ‘stored’ in Bitcoin form cannot later be released. A rehearsal of such arguments can be found in this popular blogpost.

I do see an integrity in these lines of argument. If we take the [price = value] route in judging the value of a thing to society, singling out Bitcoin’s environmental costs over the environmental costs of other things is nonsense.

Crypto Bros use a lot of electricity mining/ transacting in Bitcoin and ascribe positive subjective value to this activity even though I see this as pointless pollution with a large environmental cost.

I use electricity in connecting to the internet; I ascribe positive subjective value to this activity, but some readers of this blog might see it as pointless pollution. To one another we each say:

But I am unconvinced by the argument. I am unconvinced because I can see that such arguments would work to justify the generation of any form or level of pollution if activity of polluting delivers a monetary value. And I don’t feel comfortable signing off on that.

Maybe it’s a question about capturing the cost of the negative externalities attached to pollution properly – that would be the economist’s take, and one that I imagine Crypto folk might be happy with (although would maybe require inter-governmental alignment and enforcement that is antithetical to the Crypto crowd).

Maybe it’s just a recognition on my part that while markets and money are and have been – in my view – kind of amazing, they will always be incomplete and a poor clearing house for questions of axiological ethics. Which is a bit of a problem if they are used as such.

Money, Tax and The Left

Jo Michell has written a good piece in Tribune countering arguments from some on the Left around the necessity of taxation.

It’s a short piece that you should read, but I understood it as going a bit like this:

  • Case 1: If there is lots of economic slack in an economy (think loads of folks unemployed sitting at home wasting their skills and producing nothing), government can spend newly printed cash to boost the economy and there are frankly few immediate downside consequences.
  • Case 2: If there is not much economic slack (eg everyone has a job), printing loads of cash to spend on new projects will mean bidding folks away from their private sector jobs (which were presumably economically productive and need to be filled) and also puts more cash into the system than is wanted, so those with cash balances play a game of hot potato (trying to rid themselves of cash balances and transferring them to one another in exchange for goods, services, financial assets like shares, real assets like houses, or indeed foreign currencies – resulting in higher prices for some or all of these).
  • The problems in Case 2 can be addressed with taxation. This reduces aggregate cash balances and “the wasteful consumption of the wealthy” to make resources available for “socially-useful spending”. And actually, taxation can be deployed in Case 1 too.

Jo’s beef looks to be with the MMT crowd, who I think he reckons have dragged elements within The Left away from understanding taxes as useful (given that the magic money tree really does exist). Jo was pretty careful not to go all inflationista in his piece, but it’s plenty obvious that monetary stability – or what Keynes referred to as confidence in the currency – is at the heart of his argument.

But I do wonder whether some folks on the Left may have taken to MMT either because:

  1. they have different views around the strength of institutions in the political economy (eg, print now and we’ll have no trouble whacking taxes and/ or interest rates up at a later date if we need to). I have my doubts. Or;
  2. they see lower confidence in the currency as a *desirable outcome*.

There are lots (and lots) of problems with capitalism as an organising system, but I would not describe myself as anti-capitalist and as such see a reduction in confidence in money as an undesirable outcome.* If I were an anti-capitalist Leftist then Keynes’ famous argument (citing Lenin) that the best way to destroy the Capitalist System is to debauch the currency makes debauching the currency sound pretty interesting!

I’d always understood this Keynes/ Lenin argument to be that you need money to make capitalism function, and so destroying money as a store of value, unit of account, and ultimately as a medium of exchange through hyperinflation collapses the economic system. Pretty straightforward. If this is what attracts some on the Left to MMT I don’t think Jo’s arguments for creative use of the taxation system will shift them.

But it was interesting today to read chapter 6 of The Economic Consequences of The Peace from which Keynes/ Lenin’s phrase is taken. I’m now less sure that my straightforward understanding of Keynes/ Lenin’s argument was right.

Keynes elaborates that inflation destroysI Capitalism by:

I find this fascinating. These are four points that I would recognise as being pretty zeitgeisty, although not points I would have immediately associated with inflation. I don’t know whether they really are the Four Horsemen for the Capitalist system, but they seem to be gaining ground despite the absence of inflation.

With regard to the use of the tax system, I’m with Jo here, and have been for some time. Monetary financing needn’t spell disaster and may indeed be required to prevent deflationary spirals. But tax serves a purpose – in fact many. What sort of taxes should the Left focus on? Wealth taxes!


* I reckon that many/ most of the problems associated with capitalism can be offset or corrected by a liberal democratic State designing and revising a rules-based framework, with discretion to compensate for inequitable outcomes and the power to offset market failures of many descriptions. This makes me, I think, a social democrat. This may be naive, but I would prefer to channel energy into improving the rules/ governance rather than dismantling the construct.

A Better Fiscal Rule

Neville Hill and I originally wrote this piece in early summer 2019, but never quite finished it. As all the excitement over fiscal rules ahead of the 2019 UK General Election kicked-off, we wrote up a précis which was published by FT Alphaville.

We then submitted a longer version to the Society of Business Economists’ annual Rybzynski Prize and were delighted to be shortlisted. Since then we have received a number of requests for the longer version. Here it is:


Fiscal rules have never been so prevalent among nations, but most fiscal rules fall victim to charges of ineffectiveness, over-complexity, asymmetry and harmful pro-cyclicality. The succession of fiscal rules – national and supranational – to which the United Kingdom has made itself subject since 1992 are no exception. We argue for a new fiscal rule that might help secure fiscal credibility that is intuitive, counter-cyclical and potentially symmetrical.

Today the United Kingdom targets a falling debt stock as a percentage of GDP and a cyclically-adjusted budgetary balance over the medium-term. Eurozone member states are subject to a plethora of deficit constraints and an obligation to reduce the debt stock to 60% of GDP.

Most of these rules are being met. But these rules are no longer fit for purpose in a world where policy rates are low and central banks have little room to ease.

Why a fiscal rule?

The economic, political and social consequences of each of debt default and high inflation are severe.[1] And there appears to be a limit to the proportion of total economic output that can be directed by a government towards public debt service, bounded by political, social and financial market parameters.[2] Furthermore, while monetary sovereigns can always summon into existence more money to service debt, there are good reasons why they might not always want to.[3] As such there is a powerful imperative towards limiting or reducing debt on the part of governments seeking to maximise long-term fiscal flexibility.

But while long term fiscal sustainability is a goal upon which every government can agree, the short-term incentives to increase spending can prove politically irresistible.[4]

Fiscal rules are means by which governments can seek to bind themselves (and demonstrate to others that they are bound) to the masts of fiscal rectitude. And while mixed, the evidence suggests that episodes of significant debt reduction have been associated with the introduction and use of fiscal rules across a number of countries over many decades.[5]

Fiscal rules fall into one of four categories – budget balance rules (maximum deficit), debt rules (creating upper limits on government debt), expenditure rules (limiting growth of expenditure) and revenue rules (generally targeting revenue as a percentage of GDP). [6] Each of these approaches have sought to secure debt and, consequently, monetary sustainability. Debt rules go directly to target an upper bound for government debt stock, while the others provide intermediate targets that will restrain or reduce the level of government debt. The rules are all asymmetric – they instruct governments to cut deficits or debts, but do not set a lower bound, or instruct more expansionary policy. That may reflect the fact that they were largely designed in the wake of the fiscal and monetary policy failures of the 1960s-1980s, when there was a compelling need to demonstrate fiscal and monetary credibility.[7]

What’s the problem?

Existing fiscal frameworks recognise the prospective dangers attached to a high and spiralling stock of debt, but are not designed to see the dangers of inappropriate deleveraging. Pro-cyclical fiscal tightening (raising taxes and cutting government spending in the midst of a slowdown or recession) can be especially damaging to the economy,[8] and the rules described above, targeting maximum debt stocks or government deficits, are prone to signal the necessity of fiscal tightening at precisely the moment of maximum pain.[9] Critically, they assume that monetary policy will be free to administer short-term demand management and so may be mis-specified when rates are at or near the lower bound.[10]

And they willfully ignore information embedded in financial market prices, seeing markets only as prospective constraints to fiscal flexibility: a capricious and constant source of risk. This, as we discuss below, presents another source of asymmetry: governments cannot ignore the message of very high borrowing rates, no matter what their fiscal rules are.

Simply put, while any fiscal rule will be clunky, the existing rule set poses meaningful problems and dangers when policy rates are at or close to the lower bound, as they have been for the past decade.

A better fiscal rule

A key aim of fiscal rules is “sustainability”: to ensure public finances are managed to avoid the risk of default or inflation. And although the size of a government’s debt obviously has implications for sustainability, it is not the only metric that matters. Nor is it the most important.

Sustainability is the economic and political capacity of the state to meet its financial obligations. We think debt interest service as a percentage of GDP is a better metric than debt stock when it comes to capturing the concept of sustainability. When debt service is very high, governments find it challenging to dedicate tax revenues to paying coupons on bonds. But conversely, when it is low it would suggest there is room for fiscal expansion.

Debt service is a product not of stock alone, but of the size, tenure and type of existing debt stock, the growth in economic output, primary fiscal balance, choice of new issuance, and the level of market interest rates. These factors, furthermore, are interdependent.[11] In summary, debt service is a function of the past fiscal legacy that must be managed, the present economic environment that is ever-changing, and external expectations as to future economic and monetary conditions. Furthermore, it responds to changes in its determinants fairly slowly, especially if the debt and issuance tenure is long (see Appendix). That means fiscal rules based on debt service targets will generally only necessitate gradual changes in fiscal policy, not abrupt dislocative shifts.

Debt stock and debt service then are related, but distinct. They can move in opposite directions for decades as a glance at recent United Kingdom budgetary history can show. In 1980 the United Kingdom spent 5.2% of GDP towards servicing government debt; debt to GDP stood at 46.2%. Fast forward almost forty years and we can see debt service as a percentage of GDP having fallen to 2.7% of GDP (2.0% of GDP after the fiscal impact of QE is taken into account), while debt to GDP had almost doubled at 89% of GDP. This startling fall in debt stocks and interest costs moreover is not unique to the United Kingdom (see Figures 1 and 2).

Figures 1 & 2: Debt stocks as % of GDP since 1978; Interest costs as % of GDP since 1978

Fig 1-2

Instances where interest costs have reached 5% of GDP have tended to be met by alarm in Finance Ministries, and fiscal adjustment programmes have tended to follow (Canada’s 1985 and 1995 Plans, Italy 1988 and 1995 Plans, the UK’s 1976 IMF programme and 1980 Medium-Term Financial Strategy).[12]

There have been multiple other instances where fiscal adjustments have been delivered by G7 governments – most recently the UK’s 2012-2020 programme of austerity that seeks to target debt stock. But as Blanchard and Tashiro (2019) argue in relation to Japan – where debt-to-GDP is more than 2.5X higher that seen in the UK and debt service is 30% lower than the UK as a percentage of respective GDP – the appropriate fiscal policy for the Ministry of Finance may well be an increase rather than a reduction in primary deficits over coming years, given the macroeconomic challenges of re-anchoring inflation expectations in firmly positive territory and delivering higher levels of economic growth while the Bank of Japan remains at the effective lower bound. A higher debt stock is not a desirable end in itself, but nor is it necessarily a greater danger than an inappropriately tight fiscal policy when debt service is low.

The reason for this fall in debt costs despite the surge in debt was the collapse of market interest rates. In 1980 long term Gilt yields stood at 13.9%, and by 2017 they had fallen to 1.0% This decline in market interest rates reflects changing market expectations of economic growth and inflation. But the fall in market interest rates from a fiscal policy perspective appears to have been taken as an event rather than as a signal. A government’s decision to ignore the information embedded in asset market prices short-circuits the ability of easy monetary conditions to prompt economically stimulative fiscal policy; refocusing fiscal rules towards debt service would remove this short-circuit. Fiscal rules focused on debt service also remove an asymmetry. As we have seen time and again in the past, markets pricing extremely high rates provoke or necessitate a restrictive policy from governments. But extraordinarily low borrowing costs do not provoke the opposite – a stimulative policy – as governments are often constrained by nominal targets for the stock of their debt or the size of their deficits.

A symmetrical fiscal rule?

While a new fiscal rule that targeted a maximum debt service ratio would be a vast improvement on the current fiscal rules, it is worth exploring whether there are further benefits to be had by also setting a minimum debt service ratio. We believe that so doing would deliver three principle benefits.

Firstly, it would introduce more powerful counter-cyclical fiscal impulse. The government would be prompted to take the initiative to invest when the cost of so doing was lowest, the need was at its highest, and the potential knock-on inflationary consequences of which would be helpful rather than counterproductive.

Second, it would introduce a mechanism by which fiscal and monetary policy coordination might be enhanced. Lower interest rates – signalling expectations that economic output was far below potential – would prompt a looser fiscal policy to boost output, while higher rates would prompt fiscal retrenchment.[13] Furthermore, when the central bank undertakes large scale purchases of government debt (QE) it not only reduces interest rates in government bond markets, but government debt interest is reduced further by the central bank returning the difference between its policy rate and the yield to maturity of the government bonds it holds back to the government. At present, that reduces UK government debt service by 0.7% of GDP.

Thirdly, introducing a policy that forcefully and automatically addresses the Keynesian ‘liquidity trap’ removes downside uncertainty over the course of long-term economic growth and inflation. As such its adoption could keep interest rates and long-term bond yields at higher levels than they would otherwise be, and other UK financial assets might attract a reduced risk premium (associated with the prospect of obviating stagnation). But the symmetrical nature of the rule would also provide assurance that fiscal contraction would come onto the agenda if inflationary expectations became unanchored.

Again, we believe more research is required before a target for the lower bound of debt service is selected but would envisage that it would reasonably be around 2% of GDP. In 2018 the United Kingdom’s debt service costs amounted to 2.7% of GDP before the fiscal dividends of QE are counted and 2.0% after.

Most fiscal rules fall victim to either the charge of ineffectiveness, over-complexity or harmful pro-cyclicality. The succession of fiscal rules – national and supranational – to which the United Kingdom has made itself subject since 1992 are no exception. We believe that this proposal is worthy of examination as a potential exception to this charge sheet.


Appendix: Term of debt matters

When looking at a government’s debt dynamics we can see that term structure matters. Among developed bond markets included in the Bloomberg Barclays Global Treasury index, the UK has the longest average maturity (figure 2.1) – standing at more than twice the weighted average maturity of US government debt. Looking more closely at the UK’s refinancing requirements, we can see that they fall unevenly across time as principal and interest payments are, by their nature, somewhat lumpy (figure 2.2).

Figures 2.1 and 2.2: Weighted average maturity of developed government bond markets; Term structure of UK government debt, June 2019 – assuming RPI yoy at 3%

Fig 3-4

Bond market yields – determined by the price at which investors are willing to buy existing bonds in the secondary market, and rates of interest at which borrowers are willing to originate new borrowings – can vary meaningfully from decade to decade. The term of existing and new borrowings impacts how quickly these market-driven yields will inform fiscal sensitivities.

The relatively high term length of UK debt reduces the fiscal sensitivity to changing market expectations around the future course of the Bank of England’s Bank Rate, economic growth, and inflation (where a drop in each of these has historically corresponded to a fall in market yields).

To illustrate this let’s consider a (relatively bleak) scenario for the UK (table 2.1) over the next fifty years. In the first decade we model bond yields at levels not too distant to where they are today, sluggish nominal GDP growth, and a primary deficit (government budget balance before interest costs) of 1.5% of GDP. This scenario is consistent with a stable debt stock as a percentage of GDP, and gradually falling debt service costs.

From 2029-2039 we model a shock to bond yields, rising dramatically to 6%, nominal growth slowing to an even more derisory 3% per annum, and a larger primary deficit of 2% of GDP per annum. From 2039 onwards, we model a combination of continued derisory nominal GDP growth, a fiscal tightening such that the government runs a small primary surplus, and a fall in bond yields to 3%. The Retail Price Index (to which current inflation linked Gilts are indexed) is modeled to stay steady at a growth rate of 3% per annum.

Table 2.1: scenario to illustrate the implications of financing term

table 2

Figures 2.3 and 2.4 show how the modeled scenario can have quite differing implications for both the fiscal cost of debt service and for the contemporary stock of debt as a percentage of GDP. The purple line assumes that primary deficits and refinancing needs are met exclusively through the issuance of new five-year bonds at the modeled marginal bond yield (which is assumed to be stable across all maturities). This choice of tenor sees interest costs rise rapidly to 7% of GDP between 2029 and 2039 before receding even more rapidly to 4% over the subsequent five years. Debt to GDP rises from 2029 until 2044 when it reaches 135% before beginning a glacial decent, reaching 127% by 2071. The grey line illustrates the impact using the same assumptions but this time assuming that financing needs are met through the exclusive issuance of ten-year bonds. The light blue line shows the impact of issuing exclusively twenty-year bonds and the dark blue the impact of issuing exclusively thirty-year bonds.

Figures 2.3 and 2.4: Interest as a % of GDP; debt as % of GDP

Fig 5-6

Concerns that a fiscal rule associated with the fiscal costs of debt service rather than debt stock might put the government at the mercy of markets capture a truth that governments’ fiscal costs of debt are determined in part by market yields.

But every government seeking to finance its expenditures through the bond market will already be sensitive to market yields; the degree to which governments have heightened or dampened sensitivity to market movements is very much in their control of their debt management functions. Shifting to a fiscal rule that focused on debt service would put the question as to how much sensitivity the government wanted to have to market rates firmly in the hands of the hands of the government.


Blanchard, O. 2019. ‘Public Debt and Low Interest Rates’. American Economic Review 109, no. 4: 1197–1229

Blanchard, O & Leigh, D. 2013. ‘Growth Forecast Errors and Fiscal Multipliers’, IMF Working Paper No 13/1

Blanchard, O. & Tashiro, T. 2019. ‘Fiscal Policy Options for Japan’, PIIE Policy Brief 19-7, May 2019.

Borensztein, E. & Panizza, U. 2008. ‘The Costs of Sovereign Default’, IMF Working Paper 08/238, 2008.

Broadbent, B. 2014. ‘Monetary policy, asset prices and distribution’, speech given a the Society of Business Economists Annual Conference, October 23.

Debrun, X. & Kumar, M. S. 2007. ‘Fiscal Rules, Fiscal Councils and All That: Commitment Devices, Signalling Tools or Smokescreens?’ in: Banca d’Italia (eds.) Fiscal Policy: Current Issues and Challenges, Papers presented at the Banca d’Italia workshop held in Perugia, 29–31 March 2007, pp. 479–512.

Fischer, S. 1996. ‘Why are Central Banks Pursuing Long-Run Price Stability?’ Jackson Hole Symposium, August.

HM Treasury, 2017. Charter for Budget Responsibility: autumn 2016 update. January.

Hughes, R., Leslie, J. & Pacitti, C., 2019. ‘Britannia waives the rules?’, Resolution Foundation, October.

IMF, 2002. Assessing Sustainability, May.

IMF, 2009. ‘Fiscal Rules – Anchoring Expectations for Sustainable Public Finances’, Fiscal Affairs Department, December

IMF Fiscal Rules Dataset, 1985-2015.

Lledó, V., Sungwook, Y., Fang, X., Mbaye, S, Kim, Y. 2017. ‘Fiscal rules at a glance’, International Monetary Fund, Washington DC, March.

Mauro, P. (ed). 2011. Chipping Away at Public Debt: Sources of failure and Keys to Success in Fiscal Adjustment, John Wiley & Sons.

Mauro, P., Romeu, R., Binder, A., & Zaman, A. 2013. ‘A Modern History of Fiscal Prudence and Profligacy’, IMF Working Paper No. 13/5, International Monetary Fund, Washington, DC

Portes, J. & Wren-Lewis, S., 2014. ‘Issues in the Design of Fiscal Policy Rules’, Department of Economics Discussion Paper Series No.704. May.

Rogoff, K. 1987. ‘Equilibrium Political Budget Cycles’, NBER Working Paper 2428.

Romer, C. D., & Romer, D.H. 2019. ‘Fiscal space and the aftermath of financial crises: how it matters and why’, Brookings Papers on Economic Activity, BPEA Conference Draft’, March 7-8.

Summers, L.H. 2014. ‘US economic prospects: secular stagnation, hysteresis and the zero lower bound’, Business Economics Vol 49, No2. National Association for Business Economics.

[1] See for example Borensztein. & Panizza. (2008), Fischer (1996).

[2] See IMF (2002) on the degree to which fiscal sustainability can be understood as a political concept.

[3] While the limiting factor on private sector or foreign-currency-denominated sovereign indebtedness is creditworthiness, the degradation of which can be observed in financial instability, the limiting factor on monetary sovereign debt and deficits is the acceptability of the currency, the degradation of which is manifest in monetary instability – via currency weakness, inflation, and/ or a fracturing of the financial system.

[4] These are the so-called ‘shortsightedness’ and ‘common pool’ problems. For shortsightedness see Rogoff (1990); for ‘common pool problem’ see Debrun and Kumar (2007).

[5] Debrun and Kumar (2007), IMF (2009).

[6] See IMF Fiscal Rules Dataset (2016) for a catalogue of past and present fiscal rules at the national and supranational level, and Lledó, et al (2017) for a history of fiscal rules by country.

[7] Sovereign debt markets exist not only to finance government deficits (or rather sterilise the monetary impact of fiscal payments in excess of tax receipts), but also to provide funding benchmarks for private bond issuers, and to serve as high quality collateral for the domestic financial system. Sovereigns not requiring external financing have issued full yield curves of benchmark securities for these reasons.

[8] Blanchard & Leigh (2013).

[9] Romer & Romer (2019) examine the motivations attached to large and small-scale episodes of austerity across the OECD since 1990 and find instances where the communicated rationale for pro-cyclical austerity was misplaced fear of loss of market access.

[10] This criticism is certainly not universal, with Wren-Lewis and Portes (2017) and Hughes, Leslie & Pacitti (2019) notable attempts to incorporate monetary constraints to a fiscal rule. Towards the end of 2019 the Conservative Party announced that the UK’s fiscal framework (requiring a cyclically-adjusted deficit below 2% by 2020-21 and the delivery of a reduction in the net debt to GDP ratio each year unless the Treasury judges that there has been a ‘significant negative economic shock to the UK economy’ (HM Treasury 2017)) would be replaced by a requirement to run a balanced current budget excluding capital expenditure from 2022, a 3% of GDP limit to net investment and a 6% of tax revenues maximum for debt interest costs, beyond which government must reassess its borrowing plans with the goal of stabilizing debt stock. While a step in the right direction, this approach retains the asymmetry of previous fiscal rules.

[11] A simple shorthand for estimating the change in debt as a percentage of GDP is the equation: , where r and g represent the interest rate and growth rate respectively (in either real or nominal terms), while d and p represent the stock of debt and primary surplus (each as a proportion of GDP). Setting Δd to zero delivers a framework for achieving a stable debt to GDP ratio. Simply put, a higher rate of economic growth, a lower interest rate, and a higher primary budget surplus are all things that help to reduce debt stock.

[12] See Mauro (2012).

[13] There is a further question as to whether a symmetric fiscal rule should be set to target interest costs net of the Bank of England’s Asset Purchase Facility (APF) payments. Doing so would turbo-charge monetary-fiscal coordination by delivering to the Bank of England a mechanism to provide direct and immediate fiscal space that must be deployed by the Treasury if the lower bound of interest costs as a percentage of GDP was not to be breached. This is because coupons paid on the gilts held in the APF by the Treasury are returns to the Treasury (minus financing costs) , and as such quantitative easing: a) delivers real fiscal dividends in the form of a lower net interest cost; b) could require the Treasury to ramp up debt issuance such that the minimum debt interest payment rule was met. The degree to which such levels of coordination would threaten the Bank’s independence and institutional credibility need though to be considered in weighing this option, as would the costs of unwind.

Electoral maths for (parents of) 6yr olds

As I’ve tweeted, my six year old asks lots of questions about UK politics. And he has views.

But most recently he’s been asking how I will vote in the forthcoming General Election. I think he’s concerned that he’s getting different answers from different grown ups who he considers role models.

Beyond the whole ‘there are differing views, each legitimate’ chat, I decided to try to teach him about first past the post system as a way of explaining the concept that people don’t even always vote for the party with whom they most identify. Without an understanding of the electoral system, voting for your non-favourite party is kind of baffling (put brilliantly here by Douglas Adams).

First past the post is not tricky, but he is six. And I’m aware that I can sometimes complicate stuff. So rather than subject him to a verbal onslaught I told him that I had made a super-simple game for him that would explain. The point of this post is to share method with other parents who might be interested to convey the idea to their kids.

Imagine you are 18 and are getting ready to vote, I said. There are fourteen voters and six parties in your election. Who do you vote for? I passed him the pad of paper and told him to fill in the last box. He handed it back.

I asked him why he voted differently. He told me. That’s it.

Doing stuff for free

As I occasionally muse on Twitter, Diane Coyle’s stuff is fascinating and wonderful. If you’ve read her GDP book or any of her papers (and you ought to) I’ve nothing really to add.

But I have made a cut-out-and-keep picture of my understanding of some of the issues raised in trying to capture changes to GDP and productivity (with apologies to Diane). This is for me rather than you. It’s a bit messy, but here it is. And I like it enough to share, because prosthletise as I do, not everyone is familiar with the basic outline of these ideas.

If you’ve read this far, I expect you have one of 3 thoughts: 1) I already know this; 2) you’ve taken something simple and made it look complex, kthxbye; 3) can you explain? The rest of this blog helps people with thought number 3.

The big black line is the Production Boundary: a threshold between stuff that happens when you Do Stuff For Money (ie, captured in GDP), and when you Do Stuff For Free (ie, everything else).

Breaking the diagram into smaller pieces, let’s start with the up arrows:

When you find a new and better way to Do Stuff For Money (DSFM), this tends to be good for GDP growth and is good for the growth of GDP/ hour worked (generally thought of as productivity growth). On the diagram, this is #1 arrow on the left and captures all sorts of things from the invention of the harvester-reaper, to the roll-out of basic management training. Many if not most of the things we *all* agree about as boosting productivity do so through this arrow.

When you find a new and better way to Do Stuff For Free (DSFF), this doesn’t *directly* touch GDP growth or productivity growth. On the diagram, this is the #2 arrow on the right and captures things like the first order effects of the invention of Wikipedia or the computer language Python, learning how to cook better food, writing a blog etc. This saves us time and/ or is an improvement on what came before. It is good stuff that increases what economists would call the consumer surplus, but won’t deliver a first order boost to GDP.

There are a lot of other lines on this chart. All the other lines cross the black Production Boundary line.

When a red line crosses the production boundary from right to left it boosts GDP. When a red line goes from left to right it reduces GDP.

Famously, marrying your housekeeper reduces GDP as they pass from DSFM to DSFF (#3), even if the stuff they do doesn’t change (although in fairness it might). Divorcing your spouse and paying them to do stuff they would otherwise do for free (#4) strikes me as less conventional, but would deliver a GDP boost. What either means for aggregate GDP/ hour I guess depends upon how much you pay them per hour (as a monetary measure of their value) versus how the output per hour of the rest of the population is valued.

And now we come to some interesting bits.

Inventing and manufacturing something like a washing machine that saves us time and improves our non-market lives boosts GDP and consumer surplus (so arrow #5 is upward sloping). That is, if people generally do their own washing rather than pay someone else to so do. And so the activity of washing clothes shifts from the DSFF to the DSFM side of the Production Boundary.

And there have been some innovations like the self-scan checkout that move tasks from the DSFM side of the Production Boundary to the DSFF side (#6). Here, firms boost output/ profit per employee by substituting a combination of capital and customer labour for their own low skill labour. Unlike the washing machine, this is a slight hit to consumer surplus, getting shoppers to do the work that firms had previously paid staff to do (although maybe reduced queue-times make up for the hit; also, maybe some people enjoy scanning). So the end result is a slight decrease in consumer surplus and higher firm profit.

Online travel agents – like self-scan check outs – boost measured output per employee by shifting some of the work to the customer. But I reckon that the proposition is a cheaper outcome for the customer (#7). So GDP takes a first order hit as offline travel agencies shrink.

And lastly there is the case that Tim Harford wrote up about Microsoft Office. Tim’s mischievously makes the case that having office workers all writing their own PowerPoint presentations, doing their own expenses, etc, rather than what they are presumably experts in doing is an insult to the principle of division of labour, and a negative drag on productivity. I’ve popped this one in at #8 sloping down from DSFF to DSFM rather than just being an arrow of questionable direction sitting on the left hand of the production boundary. The logic is built on Tim’s blog: that having to do stuff in which you are not expert but which is not very taxing (ie, make bad PowerPoint presentations) may be a welcome relief from your actual work. It might even be thought of as a pretty rubbish, but fully-paid, work break. I don’t think I agree with Tim, but I 1) can’t think of another good example off the top of my head, 2) am keen to get symmetry in the diagram, 3) love Tim’s stuff.

Who cares?

Well, I think everyone actually cares about each of GDP, GDP/ hour and general welfare. But I also think that different people are incentivised to really care a lot about one angle more than another.

Government treasury departments care a lot about raising and spending money. For this, GDP is the most immediate concern. It is hard to tax stuff that is neither money flow or money stock, and GDP is money flow. Wikipedia may be awesome, but it is hard to get a cut of the welfare they deliver in order to pay nurses’ and teachers’ salaries. Things that boost GDP in some taxable way are good for governments that seek to resource spending decisions with tax.

Financial analysts and investors care about revenue, profitability, and spare capacity. For this, GDP is important, but output per employee is – I reckon – even more so. Changes in output per hour inform at the macro level the amount of spare capacity there is in an economy, and a boost in output/ hour can boost on a sustainable basis the ultimate level of GDP in a non-inflationary manner (keeping central banks away from the brake pedal). Output/ hour is also important for understanding how corporate profit margins will change. For example, if firms automate, they can produce more revenue per employee; if the cost of automation is not high this will boost profitability, helping boost firm values. And firms and investors will care a lot if some innovation (that they don’t own) shifts stuff across the production boundary. It could maybe wipe out an industry’s viability.

Citizens care about welfare improvement. We shouldn’t obsess over GDP growth if it is associated with reduced consumer surplus. That said, the correlation between measured GDP per capita and different measures of welfare is strong (maybe related to governments’ ability to intervene to raise outcomes being contingent on taxing GDP, don’t know). Have a follow of the brilliant @MaxCRoser for a plethora of great charts, many of which show that places with plenty of GDP tend to get better health outcomes. Like this chart of maternal mortality and GDP/ capita, reproduced here from

There is sooo much more to write on this, but frankly, I am poorly qualified to write it and Diane Coyle’s stuff reads much better. Also, check out the Bean Review.

Trump’s Libra/ Nature of Money Nightmare

So Trump has weighed into the nature of money chat on Twitter overnight. It had to happen sooner or later, so seems a good prompt to break my blogging holiday.

Facebook Libra looks fascinating, although to be honest I’m still trying to work out what the point of it actually is. To recap, Facebook has announced that it will launch a new cryptocurrency in 2020. Unlike Bitcoin et al it will be 100%-backed by an as yet undisclosed basket of fiat currencies. And Facebook + partners reckon they can get through the anti-money laundering regs that they’ll need to navigate so that we can all hold Libra in digital wallets and spend it on stuff on the internet. Money serves as a medium of exchange, a store of value and a unit of account. It kind of looks like Facebook could check all three boxes. Or maybe not?

The white paper makes a big play of Libra being stable, and thus filling a hole to serve underbanked folk around the world. As a portfolio manager I assumed that they equated stability to one of two things.

  1. Stable might mean non-volatile against my base currency. So the value of Libra shouldn’t jump around too much versus other stuff I call money like, for example, Japanese yen (if I’m Japanese) or South African rand (if I’m S African). But what if the yen and Rand jump around versus each other (spoiler: they do, and a lot!)? does Libra look more stable to the Japanese or to the South African Libra-holder? And, if somewhere in the middle of these two, what about to the American or Venezuelan holder? This is not clear.
  1. Or stable might mean non-volatile against the real value of a basket of domestic goods and services in my home country (this makes more sense to me). This maybe solves for my South African vs Japanese Libra-holder quandary if the Lira and yen have very high exchange-rate pass-through rates. But they don’t. In fact exchange-rate pass almost always through looks spooky-low.

So stable means something else I guess. Perhaps, as inferred in Trump’s tweet, the bar to stable is being less volatile against the USD than Bitcoin (a super low bar), or maybe it means as-stable-as-quants-can-get-it (perhaps setting the composition of reserve assets backing Libra from an optimisation that seeks minimum volatility against a user-weighted fiat currencies?). Who knows.

Why might I care? Because if a thing-trying-to-be-money is to be a store of value, its value should be well-correlated to movements in the price of the thing I want to buy. Which is one reason that central banks are so concerned with anchoring inflation expectations. Stability is required for the store of value box to be checked. And this needs to be checked to be money.

But let’s take Trump’s anxious third tweet in his thread – the one where he worries for the fate of the dollar.

In a scenario where Libra has supplanted the dollar as a medium of exchange it’s not a stretch to imagine that it really would become stable. Pastrami on rye sandwiches in New York delis could be priced the same in Libra from one day to the next. Their dollar price (if you wanted to calculate it) would fluctuate. How would the FOMC fulfil its inflation mandate? I guess that the dollar-Libra exchange rate would become one of the most important elements of monetary policy. The idea that the US might have to try to peg the dollar to a basket of currencies, the composition of which is decided by tech geeks who have decamped to Switzerland it the stuff of Trump’s nightmares.

How does Trump stop his nightmare unfolding? We get to tweet number 2:

From a finance-person perspective Libra looks like a multi currency money market ETF. But it also looks like an assault on the State’s ability to control our monetary lives.

Ill-regulated, Libra looks like a super-easy way to evade present or future capital controls (units recorded on the blockchain in Switzerland so I guess that if a country permissions its citizens to exchange their local currency for Libra, that’s it). And I can also see that it might be a way for e-commerce to once and for all rid themselves of individual countries’ fiscal reach (if transactions all take place on the Libra cyberledger, I can see that geographical fiscal jurisdiction might get tricksy). In an extreme scenario where Libra displaces a national currency as the medium of exchange I can see it leading governments to lose a major tool of macroeconomic policy as well as seigniorage.

But States really really don’t like firms playing fast and loose with people’s money. Despite what you might infer from the GFC and its aftermath, retail client money is super-highly regulated. And States would be loathe to throw away the ability to impose capital controls or tax domestic transactions at some point in the future. FinTech is a powerful thing, but so is FinReg.

Second thoughts on miniBOTs

I generally operate on the principle that if people can show me how I’ve misunderstood something I will take that as a win. If they can do this without making me feel a complete fool, that’s an unnecessary but welcome bonus.

After blogging a couple of early (fairly negative) thoughts about miniBOTs yesterday, I had a bit of feedback. What followed was a brief epiphany during which I realised I was wrong and miniBOTs could actually be an awesome idea.

The epiphany was based on the notion that if there is slack in the Italian economy (which there surely is), and that this slack could be engaged without crowding out the private sector by the government doing generally helpful-to-long-term-human-goals-sort-of-stuff (again, likely true). This could be upgrading infrastructure, developing human capital, regional/ sectoral development financing, that sort of thing. And (importantly) the financing vehicle – miniBOTS – could be used by the government in a manner that was revenue-neutral or even revenue-accretive.

It would work like this:

  • Step 1: government puts out tender for xyz contract, payable in a certain value of miniBOTs;
  • Step 2: contractors bid for and win contract, then sell miniBOTs to folks owing large back-taxes at a discount.
  • Step 3: government collects back-taxes (and/ or taxes on the grey/ black economy that now steps into the sunlight with lower effective tax rates) that were otherwise lost. So debt doesn’t actually rise.

Reckoning that miniBOTs might trade sub-par (because, you know, some other government might turn up and think the whole idea was a bit whacko and stop accepting them as taxes in lieu of euros), contractors would probably want a few more miniBOTs than plain old euros for a given job, implying a financing yield (in terms the ratio of miniBOTs to Euros, or rather discount to par).* In other words, I believe that this new good-for-Italy-using-up-slack expenditure would cost more in miniBOTs than it would otherwise have done in plain euros.

But costing more miniBOTs than euros could be okay if a) it deploys slack productively; b) it is making investments that are possible that would have been impossible without miniBOTs, and; c) would ultimately be financed by gathering back-taxes/ black economy taxes. (The folks owing back-taxes buy the miniBOTs because – now there is a discount to par value of a potentially temporary nature – it costs them less, and they seize their chance.)

So rather than being an interesting money thing, it could actually be a bit of an elaborate tax amnesty/ tax cut in which government contractors also get a little richer than they otherwise might.

However, this epiphany lasted only 53 seconds.

My mind had sort of taken a temporary leave of absence, and I’d forgotten Gresham’s Law: the idea that bad money drives out good.

In other words the miniBOTs get bought by ordinary folk at the point at which they pay their taxes (and they then deliver them to government in lieu of euros), and there is no reason why back-taxes/ black economy taxes ever get touched.

The government still has to pay more miniBOTs than euros for stuff (because otherwise why would people take them) and legit taxpayers would buy them at a price below par (because otherwise why bother). And so the whole thing becomes self-defeating as miniBOTs collect in the Italian treasury rather than among the population as a whole, who prefer to amass monetary assets in euros unless tempted by discounts to par (and hence higher implied financing rates for the state) on miniBOTs. (And the scheme introduces scope for political favours to be awarded to connected contractors who might even repay this political debt through future party funding.)

Meanwhile, the Commission would likely look at miniBOTs as debt (because selling future receivables look very much like debt, as any credit card receivable securitization analyst will tell you), and given they look to be zero-coupon, would likely see them as tradable to next put date (see footnote), which is to say having very short term structure no matter how irredeemable they profess to be.

There is a second way that miniBOTs are interesting, and that is as an asset conjured our of nothing that has value to people in such a way as to create demand for a permanent stock (eg, a currency). In such a light, the scheme looks pretty identical to the creation of a new form of Outside Money, but as a Chartalist myself I think that for it to actually properly work the state would need to require the payment of taxes in miniBOTs rather than simply accept the payment of taxes in miniBOTs. In accepting rather than requiring, we get back into Gresham’s Law problems: the miniBOTs would need to have an implicit financing cost. And because of this the state would endure a weaker fiscal position than it would otherwise have (paying more miniBOTs than it would have needed to pay euros, but receiving back an equal number of miniBOTs to the number of euros it would otherwise have received in tax revenue). If the state chose to only accept taxes in the form of miniBOTs this changes things. But I think that this then touches on the legal tender issues discussed in the previous blog.

I’m still up for being persuaded of their merits. And if I wanted to shoe-horn Italians out of the Euro without actually bothering to make and win the case with them that this was what they would be doing (stirring up a fight with Brussels, maybe demonizing them along the way to the domestic population), miniBOTs seem a reasonable thing to introduce.

Absent this motivation, I still can’t see these merits right now. There could be a decent portion of folks who reckon a tax amnesty or tax cut would be self-funding. I know there is a decent minority of the population who think that Italy should leave the Euro. Advocates of these positions should make their case without engaging in bait and switch tactics.

Chances are you’ve thought about this more than me. If you can communicate simply and coherently why miniBOTs make senses in a way that doesn’t involve leaving the Euro please blog it (in English, sorry).

* It has been put to me that miniBOTs won’t have a term (ie a date at which they will necessarily turn into euros) and so my whole blog is somewhat off-base. But if I bought an irredeemable zero coupon euro-denominated security that was puttable at par, I would consider the term to be the first put date, and then measure its yield as a function of the discount to par and distance from par. (To get a bit more technical the first put date is the date at which it is advantageous for the holder to put the securities to the issuer, and this will actually depend on the price of all of the other securities on that issuer’s yield curve as well as the put date). I can’t think why anyone else (including the Commission) would not do the same.


Update: the Bank of Italy has put out a PDF with their initial take on miniBOTs. It is the clearest thing I have read, and I was surprised and delighted that it did not contradict my understanding as conveyed in this blogpost and it’s predecessor.