Three silly charts
From Fraser Nelson's Twitter stream come the following charts, with associated twitter commentary.
There are of course going to be lots of people who think my description of these as "silly" is mistaken or incomprehensible. So let me explain.
Chart 1 compares the increase in government debt with the increase in GDP since 2010. I'm afraid this is meaningless. Debt is a stock: it is taken on at a point in time, paid down at another point in time. GDP is a flow - it is national income.
As an example, imagine you have an annual income of £50,000 and you take out a mortgage of £150,000. Your debt/income ratio has suddenly rocketed, hasn't it? Shocking. But you aren't paying that mortgage back this year. You've borrowed that money over 25 years. What is your expected income over 25 years? I'd be willing to bet it's a lot more than £50,000. In fact if you expect to work for the whole period of your mortgage and your income to increase with inflation (assumed at 2%), then your income is £1,250,000 plus compound interest at 2% over 25 years. I'm not going to work that out, but trust me it is FAR more than £150,000 plus say 5% interest compounded (probably on a reducing balance basis) over 25 years. In fact it is so good that next year you might borrow some more money to do some home improvements, and the following year you might borrow some more money to study in your spare time. Now, if you overdo this, your lenders start to get a bit worried that you might not pay them back, so as a form of insurance against potential losses, they might increase the amount they charge you for borrowing - the interest rate. I know this seems perverse, since increasing your debt service costs actually increases the chance that you will default, but it would mean that they get higher returns in the short term, which reduces their potential losses.
Of course, what you spend your debt on matters, too. If you take out a loan to fund yourself through a masters degree, then you would expect your income to increase when you complete your studies, because your degree will open doors to higher-paid jobs. So again, comparing the debt with your income THIS YEAR is meaningless. You need to compare it with your projected nominal income over the rest of your working life taking into account expected increases due to your studies.
This is true for governments too. When governments borrow, they usually borrow over a long period of time, anything from 10-30 years or even more. And some of that money will go into public investment, which should increase national income in the future. In fact governments are more complicated than you, since even consumption spending can help to increase national income. It is very, very misleading to compare government debt increases with concurrent GDP increases and deduce from it that government borrowing and spending is out of control.
Just like your lenders, if lenders to government start to get worried that they might not be paid back, they increase the interest rate, which gives them higher returns in the shorter term. The risk to governments of high debt levels is the increasing cost of debt service. But at present, the UK has the lowest borrowing costs in history. Yes, I know, interest rates can rise, but government debt is mostly fixed rate - an increase in interest rates only has a marginal effect when the debt stock is mostly of long tenor, as it is in the UK. The fact is that the UK is better able to afford its debt now than it was in the early 1980s when interest rates were 15%.
So that's why Chart 1 is silly - apart from the fact that half of it is a projection, of course. Now let's look at Chart 2.
Chart 2 is the ratio of debt/GDP, projected to 2016. Now I've already pointed out that comparing these two is meaningless, because one is a stock and the other a flow. However, government debt is usually quoted in relation to GDP. But this chart does not tell the story that Fraser Nelson wants to tell. It does not indicate that government borrowing is "out of control".
You see, Chart 2 is a ratio. And the value of the ratio depends on both the numerator (the debt) AND the denominator (GDP). An increase in the ratio of debt to income tells us nothing about either debt or income. For example, if you lose your job, your debt/income ratio heads for the moon, since you have no income but still have the debt. But if you get another job at a higher income, your debt/income ratio drops because your income has gone up, and you may also use your redundancy payment to pay off some of your debt. If you take so long to get another job that you use up your redundancy payment and are forced to increase your overdraft, then even when you get another job your debt/income ratio increases: similarly, if your new job doesn't pay as well as your old one you have a higher debt/income ratio. This is because the nominal amount of your debt does not change with your income. I know this seems obvious, but when looking at the government debt/GDP ratio people often forget about the denominator. Debt/GDP can rise more than expected not because the government is borrowing more than planned but because GDP has disappointed. The Budget 2010 projections were based upon GDP growth assumptions that have proved unattainable. No wonder debt/GDP is higher than expected.
So although Chart 2 appears sensible, it's actually as silly as Chart 1. And it is also a projection, of course.
There's another complication too, for both charts. Government debt and national income are not independent of each other. The income of government is tax revenue, and tax revenue increases with GDP. If government tax revenue increases, it can afford to borrow less. So government borrowing usually reduces as GDP rises. Conversely, when GDP falls, government loses tax revenue, so is forced to borrow more to meet spending commitments. And government spending commitments themselves increase when GDP falls, and rise when GDP increases, because of automatic adjustments in benefit payments.
Also, remember that in these two charts debt is a stock. Reduced government borrowing due to increased tax revenues does NOT mean that the debt reduces. It just doesn't increase as quickly.
And this is the problem with Chart 3. It is simply incorrect to say that Osborne is "borrowing more in five years than Labour in 13". He hasn't borrowed most of that expanse of red and pink. Over £800bn of the stock of debt existed before Osborne became Chancellor. He has, so far, added less than £400bn to the total debt stock. Even with the weird projection (why does the chart show the rate of increase of debt suddenly steepening from 2014?), he isn't going to add much more than £200bn by 2016, giving a total of about £600bn - which is pretty much the same as Labour added from 2001 (the lowest point on the chart) to 2010. And if we assume that the projection is wrong, because GDP is increasing and that will enable the government to borrow less (ONS figures already show the public sector net borrowing requirement falling), then the eventual increase in public debt may be less.
None of these charts tell us anything useful. Nowhere is the interest cost of that debt shown, or the expected tax revenues from which it will be serviced. Nowhere is the real risk of high government debt levels explained, which - as the IMF explains - is reduced ability to absorb economic shocks. All they show is that government debt is increasing. Or, alternatively, that government is producing lots more safe securities to enable its citizens to save for their retirement. And that's a good thing, isn't it?
Related reading:
Government debt isn't what you think it is - Coppola Comment
Fiscal space - IMF
Worth remembering: debt is growing WAY faster than the economy. Right now, £2 borrowed for every £1 of growth (pic) pic.twitter.com/wG8Pd45Tlj
— Fraser Nelson (@FraserNelson) October 26, 2013
For purists here's the latest, soaring debt/GDP ratio. Just as scary. pic.twitter.com/xOpJYJIA5u
— Fraser Nelson (@FraserNelson) October 26, 2013
And finally, for the record, this shows how Osborne will up national debt more in 5 yrs than Labour did in 13 yrs. pic.twitter.com/N1o1R2ZOqQ
— Fraser Nelson (@FraserNelson) October 26, 2013
There are of course going to be lots of people who think my description of these as "silly" is mistaken or incomprehensible. So let me explain.
Chart 1 compares the increase in government debt with the increase in GDP since 2010. I'm afraid this is meaningless. Debt is a stock: it is taken on at a point in time, paid down at another point in time. GDP is a flow - it is national income.
As an example, imagine you have an annual income of £50,000 and you take out a mortgage of £150,000. Your debt/income ratio has suddenly rocketed, hasn't it? Shocking. But you aren't paying that mortgage back this year. You've borrowed that money over 25 years. What is your expected income over 25 years? I'd be willing to bet it's a lot more than £50,000. In fact if you expect to work for the whole period of your mortgage and your income to increase with inflation (assumed at 2%), then your income is £1,250,000 plus compound interest at 2% over 25 years. I'm not going to work that out, but trust me it is FAR more than £150,000 plus say 5% interest compounded (probably on a reducing balance basis) over 25 years. In fact it is so good that next year you might borrow some more money to do some home improvements, and the following year you might borrow some more money to study in your spare time. Now, if you overdo this, your lenders start to get a bit worried that you might not pay them back, so as a form of insurance against potential losses, they might increase the amount they charge you for borrowing - the interest rate. I know this seems perverse, since increasing your debt service costs actually increases the chance that you will default, but it would mean that they get higher returns in the short term, which reduces their potential losses.
Of course, what you spend your debt on matters, too. If you take out a loan to fund yourself through a masters degree, then you would expect your income to increase when you complete your studies, because your degree will open doors to higher-paid jobs. So again, comparing the debt with your income THIS YEAR is meaningless. You need to compare it with your projected nominal income over the rest of your working life taking into account expected increases due to your studies.
This is true for governments too. When governments borrow, they usually borrow over a long period of time, anything from 10-30 years or even more. And some of that money will go into public investment, which should increase national income in the future. In fact governments are more complicated than you, since even consumption spending can help to increase national income. It is very, very misleading to compare government debt increases with concurrent GDP increases and deduce from it that government borrowing and spending is out of control.
Just like your lenders, if lenders to government start to get worried that they might not be paid back, they increase the interest rate, which gives them higher returns in the shorter term. The risk to governments of high debt levels is the increasing cost of debt service. But at present, the UK has the lowest borrowing costs in history. Yes, I know, interest rates can rise, but government debt is mostly fixed rate - an increase in interest rates only has a marginal effect when the debt stock is mostly of long tenor, as it is in the UK. The fact is that the UK is better able to afford its debt now than it was in the early 1980s when interest rates were 15%.
So that's why Chart 1 is silly - apart from the fact that half of it is a projection, of course. Now let's look at Chart 2.
Chart 2 is the ratio of debt/GDP, projected to 2016. Now I've already pointed out that comparing these two is meaningless, because one is a stock and the other a flow. However, government debt is usually quoted in relation to GDP. But this chart does not tell the story that Fraser Nelson wants to tell. It does not indicate that government borrowing is "out of control".
You see, Chart 2 is a ratio. And the value of the ratio depends on both the numerator (the debt) AND the denominator (GDP). An increase in the ratio of debt to income tells us nothing about either debt or income. For example, if you lose your job, your debt/income ratio heads for the moon, since you have no income but still have the debt. But if you get another job at a higher income, your debt/income ratio drops because your income has gone up, and you may also use your redundancy payment to pay off some of your debt. If you take so long to get another job that you use up your redundancy payment and are forced to increase your overdraft, then even when you get another job your debt/income ratio increases: similarly, if your new job doesn't pay as well as your old one you have a higher debt/income ratio. This is because the nominal amount of your debt does not change with your income. I know this seems obvious, but when looking at the government debt/GDP ratio people often forget about the denominator. Debt/GDP can rise more than expected not because the government is borrowing more than planned but because GDP has disappointed. The Budget 2010 projections were based upon GDP growth assumptions that have proved unattainable. No wonder debt/GDP is higher than expected.
So although Chart 2 appears sensible, it's actually as silly as Chart 1. And it is also a projection, of course.
There's another complication too, for both charts. Government debt and national income are not independent of each other. The income of government is tax revenue, and tax revenue increases with GDP. If government tax revenue increases, it can afford to borrow less. So government borrowing usually reduces as GDP rises. Conversely, when GDP falls, government loses tax revenue, so is forced to borrow more to meet spending commitments. And government spending commitments themselves increase when GDP falls, and rise when GDP increases, because of automatic adjustments in benefit payments.
Also, remember that in these two charts debt is a stock. Reduced government borrowing due to increased tax revenues does NOT mean that the debt reduces. It just doesn't increase as quickly.
And this is the problem with Chart 3. It is simply incorrect to say that Osborne is "borrowing more in five years than Labour in 13". He hasn't borrowed most of that expanse of red and pink. Over £800bn of the stock of debt existed before Osborne became Chancellor. He has, so far, added less than £400bn to the total debt stock. Even with the weird projection (why does the chart show the rate of increase of debt suddenly steepening from 2014?), he isn't going to add much more than £200bn by 2016, giving a total of about £600bn - which is pretty much the same as Labour added from 2001 (the lowest point on the chart) to 2010. And if we assume that the projection is wrong, because GDP is increasing and that will enable the government to borrow less (ONS figures already show the public sector net borrowing requirement falling), then the eventual increase in public debt may be less.
None of these charts tell us anything useful. Nowhere is the interest cost of that debt shown, or the expected tax revenues from which it will be serviced. Nowhere is the real risk of high government debt levels explained, which - as the IMF explains - is reduced ability to absorb economic shocks. All they show is that government debt is increasing. Or, alternatively, that government is producing lots more safe securities to enable its citizens to save for their retirement. And that's a good thing, isn't it?
Related reading:
Government debt isn't what you think it is - Coppola Comment
Fiscal space - IMF
Not sure I follow that, if my income remains flat (flow), but my debt levels (stock) increase it will be harder to service that new debt level so surely it's somewhat indicative.
ReplyDeleteNot necessarily. What really matters is interest rate and tenor. If you have debt of £100,000 at 5% interest over 25 years you are paying less to service your debt than £50,000 at 15% over 10 years. One of the ways of restructuring debt to make it more affordable is to extend its maturity and reduce the interest rate - as was done with Greek sovereign debt, for example. The stock of debt is still the same but the payments are reduced.
DeleteIt does make it more affordable, but at some point it has to fall due. Yes they could roll it over again but where does this end? With everyone paying interest at ultra low rates that can never be raised because they force a default as affordability plummets. Isn't this just pretend and extend as the principle isn't going to get repaid?
DeleteBen F,
DeleteRe your worries about the “principle” not being “repaid”, basically the principle is never repaid over the very long term: that is, national debts in terms of dollars or pounds have steadily grown over the last century. Though obviously there have been short periods when a bit of repayment takes place.
In contrast “nominal” considerations, national debts decline from time to time relative to GDP because those debts get reduced in real terms because of inflation, and because real GDP grows.
Stock-to-flow ratios aren't per se silly or meaningless - return on capital is a ratio of stock to flow, as is labour productivity. But comparing stocks to flows is a dangerous business, because if you don't know what you're doing (as Nelson doesn't), you're always in danger of making the mistake that he has actually made in Chart 1, which is to compare a change in a stock to a flow.
DeleteThis is a bit of a quibble as the key point here is the one you make - if you're borrowing £2 of "debt" for every £1 of "growth" then you're almost certainly doing very well, because the debt stays at £2 until you pay it back, but the growth gives you a new £1 every year (making the correct stock/flow division would tell you that the UK was getting a 50% return on marginal borrowing, which anyone can see is a pretty fantastic return on investment to be getting. This would still be dumb economics because it's wrongly aggregated, but at least the division would be giving you something at least potentially meaningful).
As I say more of a quibble than anything - I just worried that people might get the impression that comparing a stock with a flow was always wrong, rather than something like starting a sentence with a preposition - probably best avoided but sometimes necessary.
Hi Francis
ReplyDeleteAs you asked so nicely let me be the first to reply and let me concentrate on chart two as it illustrates a bugbear of mine and also because I like all Dr. Strangelove references!
Fraser Nelson has followed the standard UK line which is to follow the Office for National Statistics UK net national debt numbers. Mostly the UK media copy and pastes these into their articles. However whilst there are differences if there is an international standard it is around or similar to the Maastricht citeria which Eurostat applies to the Euro zone. This uses a measure of gross debt which has the UK at 89.6% on a Debt/GDP ratio compared to the 75.9% the ONS produced and at an earlier time period. So ironically his chart would have looked more scary if he had taken that road. A missed opportunity by him!
A side effect of this media wide acceptance of the ONS numbers and criteria is that many in the UK think we are better off compared to Europe than we actually are as we repeatedly do not compare on a like for like basis.
I find myself agreeing much more with your subsequent paragraphs on debt/GDP ratios than the statement that they are "meaningless". Your latter points that they require interpretation, nuance and some thought are correct in my opinion. We could look at debt/wealth ratios if our ability to measure wealth was not even more limited than our efforts to measure GDP.
If he wanted data showing the UK national debt to GDP ratio falling then Fraser could have looked to page 28 of the latest release on public finances but I do often wonder how many pairs of eyes take the trouble to get that far apart from mine.
I don't think he really wants to see falling debt/GDP figures, Shaun. He's too busy spinning a debt disaster story. I shall have a look though!
DeleteThe "meaningless" chart is chart 1. The others are not meaningless, but Fraser has misused them. It's particularly said that he chose to ignore the denominator in chart 2, because that is the real story: debt/GDP is much higher than predicted because we have not had the robust recovery that the OBR forecast. I am very tempted to produce a follow-up post called "Three Sensible Charts" with chart 2 plus the other two charts he tweeted!
Hmmm ... I agree with you ... mostly.
ReplyDeleteMy largest problems with the Fraser Nelsons of this world is their fixation on government borrowing. It is completely self-defeating for the private sector to complain about govt borrowing and spending when the beneficiary of this activity is always the private banking sector.
The public sector borrowing services private sector debts along with its own;
- The loans taken on by governments are always from the private sector! The complainers must think governments are borrowing from Martians.
- The private sector cannot service its own debts as service is an exponentially increasing cost, this is the bootstrapping problem. Not every firm can earn a surplus in an economy unless there is 'new money' in the form of unsecured debt added continually to the system. Only government can run perpetual deficits (borrow to service debts and not fail).
- The Fraser Nelsons believe that national governments must meet expenses and earn profits like firms. Without government deficits there would be no firms (all industrial firms are fundamentally loss-making).
- That this last is so is self-evident. If any firm could pay its own way it would do so, there would be no debts as the firm would be deployed to the degree necessary to retire all of them. Instead, the more industrial firms, the greater increase in debts.
National ledger accounting: public sector deficit = private sector surplus (wealth).
Also: GDP is measure of public sector borrowing + private sector borrowing (net investment and consumption cash flows) + net foreign exchange (more borrowing) ... GDP is debt flow tally held up to other forms of debt tally (as you point out minute-by-minute debt stocks). Indeed all the measures are increasing as there is no other way to retire debts but to borrow more from someone else!
: )
Ramify this to the situation in USA; higher debt levels but more affordable than ever.
ReplyDelete"He has, so far, added less than £400bn to the total debt stock" and with the BOE buying up £375bn with QE alone (leaving aside "Funding for Lending" and "Help to Buy" etc.) has the British state actually borrowed anything much at all ? (I'm tempted to put @Fraser Nelson because I think your blog is spot on as ever).
ReplyDeleteI wonder what the charts would look like using the "Whole of Government Accounts"
http://www.nao.org.uk/highlights/whole-of-government-accounts/
"WGA is the one place where you can see the ....total value of Quantitative Easing to Government after cancelling out the complex relationships between the Government bodies involved (Bank of England, its Asset Purchase Facility Fund, HM Treasury and the National Loans Fund)."
I'd expect more context and thought from a commentator. Many of us are concerned at the ever rising debt (public and private) and the thought of ZIRP 'forever' - even if many economists dispute we should be overly concerned. Maybe greater concern would be private debt and falling wages esp if rates rise & £ falls given energy and food need importing.
ReplyDeleteIf you're never expected to pay it back, no problem. To me the key question is, borrowed from whom (and what happens when they no longer want to lend to you)? If it's all free-as-air money created on demand by the BoE, how can this ultimately not be inflationary? If its stops, how can it not be cruelly deflationary? For the moment, the monetary inflation appears to be offset by declining velocity, but surely the latter cannot go to zero.
ReplyDeleteAnd another is, who is starved of credit while the government pigs out, and what does that do to the country's economic base?
Most government debt in Western countries is borrowed from its own citizens. It is the primary source of safe assets, e.g. for retirement saving. So the primary holders of government debt are institutional investors such as pension funds who hold it on behalf of ordinary people. The other important holders of government debt are banks, who hold it as a liquidity buffer because it is easily sold for cash in the event of a liquidity crisis. Currently the biggest holder of UK government debt, however, is the central bank, because of asset purchases (QE).
DeleteI don't want to explain here why QE is not, and never will be, inflationary. I suggest you visit the QE Debate site (link at the top of this screen), which has a whole discussion thread with numerous links on the subject of the inflationary (or possibly deflationary) effect of QE. It is not relevant to the discussion here.
I don't really accept that government debt crowds out private investment. Because it is a "safe asset" it serves a different purpose. It enables those who can't or don't want to put their money at risk to save safely over the longer-term, and it is a portfolio hedge asset. This last actually means that to some extent it ENCOURAGES private sector investment, since within a portfolio it reduces the risk of risky assets. The risk reducing effect of long-term government debt is not taken seriously enough in economic circles, and nor is the use of short-term debt as collateral in financial markets. Forcing investors to take risk they don't really want is destabilising.
"I don't really accept that government debt crowds out private investment."
DeleteCheck out Spain and Italy's loan growth and how much government bonds have been bought with LTRO money and then maybe, just maybe you can somewhat accept some kind of "crowding-out" at play.
"Forcing investors to take risk they don't really want is destabilising."
That's exactly what Global ZIRP is doing and leading to mis-allocation of capital on a very large scale.
Best,
Martin
Martin T,
DeleteIf government borrowing does initially crowd out private investment (presumably because government borrowing raises interest rates) the central bank will simply counter that by stopping interest rates rising. Or the CB may even cut rates.
The only circumstance where the CB wouldn’t do that is where the CB thought government was wrong to implement fiscal stimulus (aka borrow and spend).
Obviously I’m talking about a country that issues its own currency here, and that’s what Frances’s above article is about: the UK in particular. Spain and Italy, to which you refer, are in the Eurozone and are a different kettle of fish.
Re your claim that ZIRP leads to a “mis-allocation of capital”, you’re assuming that because a significant positive rate of interest has been paid on government debt in recent decades, that therefor that is the natural order of things, and the current near zero rates are an aberration.
Warren Mosler has argued that in fact if we wanted the natural or free market rate of interest to prevail, the rate of interest on government debt would be zero. See:
http://moslereconomics.com/wp-content/graphs/2009/07/natural-rate-is-zero.PDF
Also Milton Friedman argued for a regime in which government didn’t borrow at all: that is, the only liability issued by the government / central bank machine is money (monetary base to be exact). And monetary base normally pays no interest.
I agree with Mosler and Friedman. But this is a complicated issue, and one cannot do it justice in a comment on a blog like this.
Martin.
DeleteLending to businesses in Spain and Italy is constrained by very high interest rates, because of market perception that these countries are in economic difficulty. It is certainly not due to "crowding out" by government debt.
LTRO money was used to buy government debt which is then used as collateral for repo funding from the ECB. It was a backdoor way of bailing out both distressed sovereigns and their banks. The "doom loop" is the fact that as a result, banks and sovereigns are now propping each other up. But without the LTRO, they would both have collapsed. Again, this has nothing to do with "crowding out". Spanish and Italian banks are certainly not healthy beasts capable of taking on the risk of lending to the private sector. They are, frankly, zombies.
It is not ZIRP that is forcing investors to take on more risk than they want. It is scarcity of safe assets - partly due to QE.
Ralph,
DeleteAs you say this is a complicated issue and not the place to discuss it. But I must correct you one one point. Most of the monetary base DOES bear interest. It is only physical cash that doesn't. Central bank deposit facilities are interest-bearing. Currently the Bank of England is paying zero, but in more normal times it pays bank rate on deposits.
"Re your claim that ZIRP leads to a “mis-allocation of capital”, you’re assuming that because a significant positive rate of interest has been paid on government debt in recent decades, that therefore that is the natural order of things, and the current near zero rates are an aberration."
DeleteDear Ralph,
Negative interest rates have ALWAYS led to mis-allocation of capital and bubbles what is called the "Cantillon Effect". On that subject, I suggest you read again both Wicksell and Irving Fisher.
Classical equation of exchange, MV = PQ, also known as the quantity theory of money. Quick refresher: PQ = nominal GDP, Q = real GDP, P = inflation/deflation, M = money supply, and V = velocity of money.
-Endogenous money, PQ => MV (Hume, Wicksell, Marx)
-Exogenous money, MV => PQ (Keynes, Monetarist)
In our "Cantillon Effect", we get:
Δ M => Δ Asset Prices
Abnormal low rates do not create growth, they lead to massive mis-allocation of capital and from there to a more volatile and slower growth rate. If you keep ZIRP you get a rising stock market led by multiple expansion, so you get higher and higher PEs.
When the cost of money is "too low", it pays to buy "stuff" on a leveraged basis, so you get booming housing prices in London and California back to good old house "flipping" fashion. That's investing in un-productive asset so clear"mis-allocation" of capital unless you think that a "house" is a productive asset.
"Even apart from the instability due to speculation, there is the instability due to the characteristic of human nature that a large proportion of our positive activities depend on spontaneous optimism rather than mathematical expectations, whether moral or hedonistic or economic. Most, probably, of our decisions to do something positive, the full consequences of which will be drawn out over many days to come, can only be taken as the result of animal spirits—a spontaneous urge to action rather than inaction, and not as the outcome of a weighted average of quantitative benefits multiplied by quantitative probabilities." - John Maynard Keynes, The General Theory of Employment, Interest and Money, 1936.
Even Keynes new about the mis-allocation risk due to mis-pricing of interest rate by a central bank.
In similar fashion to the Cantillon Effects, which led to the Austrian Business Cycle Theory and the role of "exogenous" money, Keynes held the view that "animal spirits" lead to damaging speculation and he was a proponent for government restrictions on investment to avoid the formation of "asset bubbles".
But, the "behavioral psychologist" in me would argue that large speculative episodes throughout histories are not engendered solely by human nature. Multiple examples of boom, and bust cycles have shown that the "Cantillon Effect" leading to the formation of asset bubbles shared common trait of "stimulation". For instance the housing bubble in the US was clearly the result of government policy in conjunction with increases in the money of supply with incorrect interest rate levels leading to speculation and "mis-allocation" of capital.
When it comes to inflation, the "Cantillon Effect" is another way of looking at inflationary pressures, but of a different nature. Inflationary pressures are not always strictly tied up to consumer prices, but, can be transmitted first in asset prices. While inflation in the US peaked at 5.6% in August 2008 prior to the stock market crash of October, the "Cantillon Effects" witnessed in the surge of assets prices were largely ignored by the US central bank.
Best,
Martin
Frances,
DeletePractice in central banks varies, but the first link below says the Fed didn’t pay interest on reserves till 2008: see first paragraph.
http://www.frbsf.org/education/publications/doctor-econ/2013/march/federal-reserve-interest-balances-reserves
First sentence here backs that up:
http://scholar.harvard.edu/files/stein/files/optimalconduct.pdf
First few sentences at the link below say that Fed and Bank of England have only paid interest on reserves since the crisis began, whereas the ECB has paid that interest since 1999.
http://seekingalpha.com/article/1726912-the-good-the-bad-and-the-ugly-of-the-excess-reserve-policy-of-the-fed
I use the term “reserves” above, whereas you refer to “deposits” (at the BoE) and "monetary base". But they are all the same thing aren’t they (except in that some monetary base consists of physical cash)?
No, they aren't the same. M0 (the monetary base) consists of bank reserves (required and excess) and notes and coins in circulation.
DeleteThe Fed did not pay interest on reserves (required or excess) until 2008. But other central banks did. As you say, the ECB has paid interest on deposits since 1999. And the Seeking Alpha link is incorrect about the Bank of England. Before the crisis, the Bank of England paid bank rate on reserves held under its voluntary reserve scheme, which was the equivalent of the Fed's required reserves. The voluntary reserve scheme is currently in abeyance and the Bank of England is paying zero on deposit balances. But as I'm sure you are aware, an interest-bearing account may have a zero interest rate but it is still interest-bearing. Most of the monetary base is therefore interest-bearing even if it is not currently earning any interest.
Martin,
DeleteYour argument hangs on your assertion that current interest rates are too low. However, the economics profession is by no means in agreement about this. Many regard interest rates as too HIGH, not too low, because the zero lower bound prevents them falling further.
If you are correct that interest rates are too low, then yes, we would expect misallocation of capital as you describe. But the arguments of others - that interest rates are perhaps too high, or maybe just right - deserve respect too. You must support your primary assertion, rather than simply assert and then explain the consequences.
"However, the economics profession is by no means in agreement about this."
DeleteDear Frances,
"Many regards interest too high but many regards them as being too low."
In April 2011, Charles Plosser, the head of Philadelpha Federal Reserve Bank, argued that the Fed should have increased short-term interest rates to 2.5% according to the Taylor rule.
Based on the current output gap and the Fed’s economic projections, the Taylor Rule would suggest that the Fed’s ZIRP should continue only until early 2014. But you can expect once again the Fed to be behind the curve as they have always been.
John Taylor, an economist at Stanford University, published the formula in 1993. It signals the Fed’s benchmark should be 0.65 percent, or 40 basis points above the upper range of the current target interest rate for overnight loans between banks, assuming an inflation of 1.7 percent, unemployment of 7.9 percent and a nonaccelerating inflation rate of unemployment, or NAIRU, of 5 percent. NAIRU is the lowest unemployment rate an economy can sustain without spurring inflation.
But there is an issue with NAIRU. The NAIRU analysis is especially problematic if the Phillips curve displays hysteresis, that is, if episodes of high unemployment raise the NAIRU. This happens when unemployed workers lose skills so that employers prefer to bid up of the wages of existing workers when demand increases, rather than hiring the unemployed.
"You must support your primary assertion":
Here it is again:
"Cantillon Effect", we get:
Δ M => Δ Asset Prices
How is that not supporting my argument?
Share prices, for example, must ultimately be justified by earnings. But while equity prices have been rising in the US, the economy remains sluggish and the outlook for corporate profits is not particularly "bright".
Are you familiar with Wicksell's work?
http://cameroneconomics.com/wicksell.pdf
It would be nice to see you being more specific as well rather than making a vague comment such as "the economics profession is by no means in agreement with this". Can you define the "economics profession" for me?
Best,
Martin
Martin,
DeleteYes, of course I'm familiar with Wicksell's work. And I'm also familiar with the Taylor rule - and its problems. As far as I can see - and you yourself acknowledge in your comment - present fundamentals in the US do not justify higher interest rates. You are in effect suggesting that interest rates should be raised now despite this, because you think the Fed is relying too much on lagging indicators. My point is that if you think interest rates should be raised now, you really need to provide the evidence on which you base this judgement. Simply saying "the Fed is always behind the curve" doesn't justify an action that is not consistent with present fundamentals.
I'm not about to define "the economics profession". I'm sure you and I would not agree on its dimensions. However, there are a number of economists who have suggested that interest rates should in fact be negative and are trying to find ways of eliminating the zero lower bound problem caused by the existence of non-interest bearing physical cash. Have a look at the work of Miles Kimball of the University of Michigan, for example.
No I don't suggest that interest rates should rise now. That is not what I have written. I don't mind a doubling of QE provided interest rates are close to or around GDP growth in true Wicksellian fashion. That is my view.
DeleteThe Fed's dual mandate is inflation and unemployment, not NGDP. At present core PCE is well below target and unemployment well above. There is therefore no justification for higher interest rates even though NGDP is higher.
DeleteI do mind a doubling of QE, actually. I am of the view that it is serving little useful purpose and is having perverse and possibly damaging effects.
So QE is damaging and not ZIRP? I find the Fed dual mandate very stupid, you either target inflation or unemployment but I do not think you can efficiently target both. Volcker did not care about the unemployment mandate and did manage to stabilise the US and break inflation. Of course republicans and democrats members of Congress were fuming because Volcker preferred a short painful recession path.
DeleteMartin,
DeleteI'm not going to discuss here exactly why I think QE is damaging. I've written about it extensively myself and collected contributions from others on the sub-site "The QE Debate", the link for which is at the top of this screen. Suffice it to say that the effects of QE are poorly understood. It appears to be effective as a crisis measure to prop up asset prices, but far less effective as a measure to counter economic stagnation and fiscal contraction. And it has unintended and possibly damaging consequences in financial markets, emerging markets and the wider economy.
You are clearly not impressed with the dual mandate and the Taylor rule, but you have not explained what you would have as an alternative. Are you suggesting that the Fed should adopt NGDP targeting?
No i am not suggesting NGDP targeting either. Did I wrote this? No. What would I have as alternative as the Fed is dangerously flirting with deflation is Fisher's solution. Forward Tax Receipts.
DeleteBest,
Martin
Ok Martin, I've read your own post on this now.
DeleteForward tax receipts are of course fiscal policy, not monetary - if they are issued by the Fed then they are monetization of government debt. This doesn't cause me a problem - I'm generally a fan of fiscal policy as a solution to deflation (which we would have if the Fed wasn't propping everything up). But it would be extremely difficult to implement given the present state of US politics. I can't see a Republican-dominated Congress agreeing to this in a million years, can you?
It still doesn't answer my question as to what you think the Fed's mandate should be, though.
Hi Frances, the Fed's mandate should be what the one of price stability. Adding full employment to its mandate in 1978 was a political stunt to make the Fed much more a political tool. Volcker did not fall prey to this trap which had been set up by the Congress. Furthermore, I quite like the introduction of macro prudential measures within a mandate of a central bank such as the ones the Central Bank of Sweden is thinking of. Countercyclical measures such as forcing banks to put more money aside in the good times of the credit cycle are in my own opinion highly interesting.
DeleteBest regards,
Martin
Fair enough. But core PCE in the US is currently 1.2% against a target of 2%, and other measures are below target too. So even if you ignore 7.8% unemployment, there is still no justification for interest rate rises at the moment. If anything, such low inflation suggests that interest rates are too high.
DeleteI like the idea of countercyclical capital buffers too. But note they are CAPITAL buffers, which is not the same as "setting money aside". Bank capital is shareholders' funds, principally.
I would contest that point.
ReplyDeleteThe Bank of England only started to pay interest of 0.5% on reserves on 5 March 2009, which it is still doing.
Not true, I'm afraid. The Bank of England paid bank rate on voluntary reserves until 2009, when the voluntary reserve scheme was suspended due to the presence of excess reserves in the system. Since then it has paid zero on deposit balances.
DeleteI assume you got your information from the Bank of England's website? It is somewhat out of date.
Martin,
ReplyDeleteI quite agree that ideally interest rates would be left to market forces: i.e. government would not interfere with interest rates. In fact I did a post on my blog some time ago attacking interest rate manipulation:
http://ralphanomics.blogspot.co.uk/2012/03/sixteen-reasons-why-mmt-is-right-on.html
However, it’s not clear to me that interest rates are currently far below free market rates. One reason is that governments for decades and indeed centuries have borrowed not just to fund investment, but also to fund CURRENT spending. The latter is a nonsense, and it will artificially RAISE interest rates.
Second, what has happened since the crisis is that plonkers in authority haven’t dared implement enough fiscal stimulus (particularly in the US with its debt ceiling and all that), because they have a phobia about deficits and national debts. So they’ve gone for bizarre forms of MONETARY stimulus instead: ZIRP and QE, etc. That will have had the OPPOSITE effect: artificially reducing interest rates.
I’m not sure which effect predominates, but my hunch (and that’s all it is) is that the total of government borrowing still exceeds government assets. That’s supported by figures at the link below which give US Federal debt in 2011 as $12.26trillion, and Federal government assets as $3.53trillion. (But I suspect the latter figure is too low.)
http://en.wikipedia.org/wiki/Financial_position_of_the_United_States
Dear Ralph,
DeleteThe issue is not QE, the issue for mis-allocation is ZIRP.
Read Wicksell and Fisher and read my comment on the Taylor rule in my response to Frances.
Best,
Martin
Frances
ReplyDeletehttp://www.bankofengland.co.uk/publications/Pages/news/2013/010.aspx
"The Bank of England's Monetary Policy Committee at its meeting on 9 October (2013) voted to maintain the official Bank Rate paid on commercial bank reserves at 0.5%."
I did not notice any announcement in the last 10 days to change that to zero.
Bob.
Yes, that's what I said. The Bank of England pays bank rate on voluntary reserves. But it pays zero on deposit balances. See paragraph 58 here:
Deletehttp://www.bankofengland.co.uk/markets/Documents/money/publications/redbookosf.pdf
Frances, thanks now I understand what you are referring to.
DeleteEssentially only institutions without reserve deposit facilities (those guys will be getting the 0.5% interest) will be using the Operational Standing Facilities to allow friction free settlement.
From the BoE website it does appear that OSF has not had any use since 0.5% was paid on reserves in March 2009.
http://www.bankofengland.co.uk/boeapps/iadb/fromshowcolumns.asp?Travel=NIxSTxTDxSUx&FromSeries=1&ToSeries=50&DAT=RNG&FD=1&FM=Jan&FY=2000&TD=28&TM=Oct&TY=2013&VFD=Y&html.x=21&html.y=13&CSVF=TT&C=JOW&C=JOY&Filter=N
Bob
This comment has been removed by the author.
ReplyDeleteGraph 1 is not a stock of money, it is a flow of money through the treasury.
ReplyDeleteIt is labelled "cumulative increases in ... public debt" .
The subject is not the outstanding debt , the subject is the act of and the size of, the ongoing act of borrowing.
No, Dinero. A CUMULATIVE increase is a stock. If you want a flow, you need an annualised increase (assuming that you are using annual GDP).
DeleteIn this context a stock would be the past tense "accumulated debt"
ReplyDeletebut here the focus is different and plural "cumulative increases in debt - a flow of increases .
Agreed the wording is unclear especially the "cumulative increases in Gdp" bit, but the topic is the flow of increases , the annual borrowing itself - a flow.
But you are still comparing a stock with a flow. GDP is never a stock, but debt increase becomes a stock. It's not remotely meaningful to compare them in this way. If you produced a chart showing the change in the DEFICIT versus change in GDP year-on-year, that would be a meaningful comparison. But comparing change in GDP with the cumulative increase in debt is comparing apples and pears. Essentially, this chart confuses debt (a stock) with deficit (a flow).
DeleteGovernment spending borrowed money is included as a component in the GDP calculation, which makes the chart even more remarkable. If government claims to be borrowing to invest then I agree the returns, growth in GDP, would not necessarily happen in the same year.
Deleteanalyzing stock and flow at the same time , a bank manager does that when they compare a salary with a principle in a mortgage. Also Frances has an example of a mortgage in the post.
DeleteMartin T,
ReplyDeleteYou claim the Fed cannot target inflation and unemployment. Why on Earth not? If the Fed (or those implementing fiscal stimulus come to that) aim to maximise numbers employed in as far as that’s compatible with acceptable inflation, then they are ipso fact targeting “both”.
Next, you claim that negative interest rates (I assume you mean negative REAL interest rates) “have always lead to” . . “the Cantillon effect”. That idea is flatly contradicted by the experience of the last few years.
That is, we’ve had negative real rates for some years now, and the excess inflation that is part and parcel of any excess Cantillon effect is nowhere to be seen.
I use the word “excess” because the Cantillon effect exists ALL THE TIME, in that inflation is with us pretty much all the time. So just to be accurate, it’s “excess Cantillon effect” we are talking about here, rather than “Cantillon effect”.
Dear Ralph,
DeleteVolcker knew it was stupid to target both and he chose not too although it was in his remit since 1978. NAIRU is the lowest unemployment rate an economy can sustain without spurring inflation.Do we agree on that or not?
The NAIRU analysis is especially problematic if the Phillips curve displays hysteresis, that is, if episodes of high unemployment raise the NAIRU. This happens when unemployed workers lose skills so that employers prefer to bid up of the wages of existing workers when demand increases, rather than hiring the unemployed.
So please explain me how on earth can the Fed maximise employment with hysterisis being a strong headwind?
The Fed is continuing on a "wrong" path and ignoring basic relationship such as Okun's law and the prolonged negative effects of ZIRP on the labor force (capital being mis-priced, it is mis-allocated to speculative purposes rather than productive purposes), so capital is hoarded and not deployed.
Okun's law (named after Arthur Melvin Okun, who proposed the relationship in 1962.is an empirically observed relationship relating unemployment to losses in a country's production.
No inflation is not with us all the time, this is complete nonsense and you are not backing up your argument with any examples. Capitalism is by definition deflationary due to productivity gains à la Schumpeter. On that subject I recommend you read Jean Fourastié - Pourquoi les prix baissent, 1984 (unfortunately in French).
Changes in the money supply and credit have important impacts on the economy and prices and that is very basic. I suggest you have a look at Ray Dalio's video on how the economy works on youtube.com.
“the Cantillon effect”. That idea is flatly contradicted by the experience of the last few years."
Oh yeah? Please explain or prove your assertion.
I don't know if you are familiar with Didier Sornette's work but, maybe you just want to have a look at his work.
Regards,
Martin
The thing that bugs me about stuff like this is that when people like Fraser Nelson exploit the public's poor level of understanding about the economy and fiscal matters, they also help to perpetuate it. When ordinary people see tweets like this, they are reassured that their own primitive intuitive notions about economics are shared by what Krugman calls "Serious People".
ReplyDeleteI'm a Briton living in the US and when I see Internet comments I am embarrassed that fellow Britons embrace economic errors, such as the Household fallacy, or a confusion of levels and gradients, as obvious truth. In the US, you have to go to the Tea Party and the Gold Nuts to find the kind of Economic ignorance that is everyday fare in the UK.
The Cantillon effect is an ideological statement. For a start, I'm not convinced that the decision between two similar investments is very much different at 5% as it is at 3% or at 0%. You ought to be indifferent between two projects with similar profitability with regard to the cost of capital. If you're not, you don't believe their profitability is similar. Therefore, how do low interest rates cause misallocation between sectors?
ReplyDeleteSecondly, a Cantillon effect requires that some "bad" projects get greenlit just because the capital is available. This I can believe - it's blindingly obvious. Every property bubble etches it on the landscape.
Somehow, though, the effect only works in one direction. Bad projects get cleared because credit is too loose. But, apparently, good ones are never killed because credit is too tight. This was, in fact, the normal operating mode of the gold standard, which Cantillon-ists want to go back to. It was also the normal operating mode of the neoliberal central bank up to September 2007 - it wasn't that long ago that central banks were primarily credit-restricting institutions.
The statement that changes in credit availability have effects on the real economy is, I think, so true it takes a greatly respected economics professor to deny it. It is worth pointing out, though, that believers in Cantillon effects argue that they do not, except when they want them to.
The statement that changes in credit availability might have effects on all kinds of sectors of the economy, without distinction, seems obvious. The statement that changes in credit availability might have effects that are very different on different sectors, for exogenous reasons, seems sensible.
But the Cantillon effect is a statement that changes in the availability of credit have large and very different and non-obviously distributed effects across sectors, whose net effect is always and everywhere disastrous, but only when the changes are positive. Cantillon believers hold that it is possible to be granted a bank loan mistakenly but not to be refused credit mistakenly.
To put it another way, errors in the direction of the rentier interest do not exist. Errors in the opposite sense do. Now, you can set up a straw-man argument and say "well, what do you want?" but then this is only a straw man. Empiricists are aware that error is possible in both directions, and I would point out that I've not heard of anyone who thinks credit cannot overexpand. I think this points at the clown here.
Dear Alex,
Delete"The Cantillon effect is an ideological statement."
If that is the way you are arguing without using facts to disprove it, then Alex, you are indeed on a good start.
"For a start, I'm not convinced that the decision between two similar investments is very much different at 5% as it is at 3% or at 0%."
I wonder then why people doing basic finance mathematics compute IRR on their HP12 calculator.
Nota bene: The internal rate of return (IRR) or economic rate of return (ERR) is a rate of return used in capital budgeting to measure and compare the profitability of investments. It is also called the discounted cash flow rate of return (DCFROR) or the rate of return (ROR).
So yes why bother?
"how do low interest rates cause misallocation between sectors?"
Please read again Keynes, he knew a thing or two about "speculation" and the famous "animal spirits". Or what about carry trades? Credit growth has been contracting at a rapide pace in Southern Europe while "whatever it takes" from Draghi has led to peripheral banks adding $360 billion of government bonds to their balance sheet for the "nice carry".
"But, apparently, good ones are never killed because credit is too tight."
Do you have facts on that? Or that's just hearsay?
"But the Cantillon effect is a statement that changes in the availability of credit have large and very different and non-obviously distributed effects across sectors, whose net effect is always and everywhere disastrous,"
It is not disastrous, it is simply called a credit cycle, just watch the video from Ray Dalio, about how the economy works. By the way Ray Dalio just manages $150 billion with a "decent" track record, so I guess the guy understands a little bit credit cycle, "cantillon effect" and the economy.
"Cantillon believers hold that it is possible to be granted a bank loan mistakenly but not to be refused credit mistakenly."
Subprime? Does it works for you?
"errors in the direction of the rentier interest do not exist."
Wow, that's a bold statement, can you expand on this please.
" I think this points at the clown here."
I agree you can disagree but, that doesn't give you the right to be insulting, that really reflects poorly on your arguments and put you in a difficult position of invalidating most of your well constructed arguments (this is called irony and is much more subtile).
Best,
Martin
I wonder then why people doing basic finance mathematics compute IRR on their HP12 calculator.
DeleteBecause you've missed the point. Consider two similarly sized investments with similar returns (and hence, risks). Whether the interest rate is 3% or 5% shouldn't change my decision to allocate investment between them. How could it? It might well change my decision to invest at all - but then that's just a statement that monetary policy can affect the economy.
If changing the interest rate doesn't affect allocation between sectors, it doesn't lead to misallocation. (It might have interesting effects on allocation between asset-classes, but that's beside the point.)
Subprime? Does it works for you?
Depression; does it work for you? As I say, the really weird issue is that it's apparently only a problem in so far as somebody might get a pay rise. The core assumption is that it's only inflation that causes misallocation. Deflation is always good, and the only arguments I can see for that is either a) that it's not really an argument, rather a general aesthetic/emotional question of taste or b) that deflation is a fundamentally conservative force in society and that's what's actually doing the work in the whole project of Austrian economics.
After all, the notion of a structural misallocation towards investment in general is a fundamentally odd one. It's not a belief in saving, because what is saved is after all invested and the same people who will tell you this are also the most committed to perfect markets and Say's law. Is it a suspicion of growth in general? Actually, I think it's a belief that both growth and inflation are forces for change in society and that this is a bad thing.
I could go on to argue that this is the trauma of the end of the dual monarchy, and a sort of guilt complex working itself out on the part of people who would have laughed themselves hoarse at the reactionaries who thought the emperor was God's anointed and who now didn't like the future one little bit - especially as those reactionaries would have despised most of the gang for being Jewish.
I think you can trace a history of ordoliberal thinking that goes "things were OK under the kaiser, so the problem must be that we got rid of the authoritarian state and the socialists unleashed the smelly mob, so clearly an authoritarian economic order is a prerequisite for a liberal society such as I myself want to live in".
The Keynesian political worldview is the opposite - if society does not respond to the economic aspirations of the masses and specifically their entirely justified desire for security, the masses will damn well make it do so, and you might not like the other options (communism, fascism, development dictatorship, Maoism), especially if you're a deeply Europeanised, arty farty gay professor from Cambridge on first name terms with most of the cabinet and therefore not exactly a great life insurance risk in a revolutionary situation either.
Dear Alex,
DeleteAlthough I find some genuine interest in exchanging and debating with Frances and Ralph, I must admit I have a hard time enjoying debating with you to be brutally honest. Take for instance your latest reply, I find it very confusing and for me it is very difficult to grasp the point you are trying to make. So I guess I am not "intelligent"enough to pursue debating with you and I will call it a day.
Have a nice day,
Best,
Martin
There was a comment earlier about the difference between ONS figures for UK deficit and debt and Maastricht numbers.
ReplyDeleteIn fact, the ONS publish both sets of numbers, and under the Maastricht definition UK government gross consolidate stands at 88.3% of GDP up from 85.0%.
Incidentally, although I am pointing this out, I don't think it is in the least important. What should concern us is the direction and pace of fiscal consolidation in the UK, not comparisons with Europe.
In fact, having, or supposedly having, lower debt levels than Europe isn't a matter of being "better off" at all. It's just a question of the balance of transfers between public and private sector, or vice versa.
And there is a much better argument than debt comparisons that we are better off than Europe, and it is that we have our own currency and are not subject to misguided eurozone austerity policies.
It's not just a question of the balance of transfers between public and private sector it is a transfer to the recipients of goverment spending from future tax payers
Delete