skip to Main Content

World Bank boss has a brain attack

The World Bank boss Robert Zoellick claims that we should all return to the Gold Standard to restore economic stability in the World economy. He is crazy. Sorry! The G-20 meeting in Seoul this week will obviously be concentrating on side issues such as the impact of the latest US quantitative easing plans on world inflation and the international currency system which many commentators are now claiming is in turmoil. Zoellick’s proposal will be added to the agenda which will reinforce what a waste of time these meetings are turning out to be. Zoellick’s call for a gold standard is just another one of these conservative smokescreens that attempt to solve the problem by denying it. They are all just expressions of obsessive and moribund fear of fiscal policy and the erroneous allegation that budget deficits cause inflation. So we will get a G-20 communiqué in a few days calling for more international cooperation in trade and currency settings and more fiscal consolidation and the need for on-going discussions about the creation of a new international reserve currency (perhaps a gold standard). But all these words will be in spite of the real policy agenda that is required – more public spending. What will they come up with next?

In a recent Op-ed (November 8, 2010) – The G20 must look beyond Bretton Woods – Zoellick said among other things that:

… the G20 should … build a co-operative monetary system that reflects emerging economic conditions. This new system is likely to need to involve the dollar, the euro, the yen, the pound and a renminbi that moves towards internationalisation and then an open capital account.

The system should also consider employing gold as an international reference point of market expectations about inflation, deflation and future currency values. Although textbooks may view gold as the old money, markets are using gold as an alternative monetary asset today.

Why would a return to the Gold Standard be crazy?

Remember as you read on to keep reminding yourself of the obvious – a sovereign nation with a flexible exchange rate can also use fiscal policy to maximise domestic outcomes if they desire notwithstanding external constraints (and domestic constraints like lack of food!).

Zoellick’s proposal is just one of a long line of proposals that get oxygen because governments are scared to use the tools that they already have and which work best when there is an floating exchange rate.

So we get Ireland destroying itself because it gave up both its currency sovereignty and its capacity to float a sovereign currency when it joined the Euro. You might like to read the latest insights from Irish academic Morgan Kelly who presents a dire update on the situation in the nation that the Euro bosses held out as the demonstration state both in the lead up to the crisis and subsequently in embracing austerity the earliest. The nation will collapse before long.

You also see central banks trying once again to revive their flagging economies with quantitative easing – again. This is despite the fact that banks do not lend reserves; US corporations are cashed up to the hilt; and US households are embracing a new period of frugality because they have huge debt overhangs that they know have to be extinguished. As I explained in yesterday’s blog – Religious persecution continues – quantitative easing will not be the silver bullet. So why are they doing it again? Answer: because the US government refuses to implement the appropriate fiscal stimulus.

All the talk about quantitative easing flooding the world with US dollars and creating an inevitable inflation are just hot air.

And that segues in to the largely irrelevant debate about the need to rebalance global trade patterns from which the call for some return to a fixed exchange rate system materialises. If you want to see how a fixed exchange rate system kills nations you just have to look at what is happening in Greece, Ireland and Spain at present.

Trade imbalances (deficits) have to addressed via domestic contraction which biased most economies to permanent states of stagnation – high unemployment and flagging living standards.

Please read my blogs – Twin deficits – another mainstream mythYuan appreciation – just another sideshow – for more analysis of that sideshow.

So it is just another part of the anti-fiscal policy agenda that has resurfaced and gathering pace even when the effects of too little fiscal intervention are staring us in the face and have been doing so for nearly 3 years now.

In part, the anti-fiscal policy agenda is sourced in a deep anxiety about inflation. The contradictions in this stance are stark but that never stops the proponents pushing it. Spending equals income whether it is private or public. The composition of the output that creates the income might be different but you need spending to create income.

Too much spending whether it be public or private will be inflationary. There is not something magical about public spending that makes its more prone to generating inflation. Further, as I have explained before it is the act of spending which introduces an inflation risk (depending on the state of capacity utilisation) and the monetary operations that accompany the spending (for example, debt-issuance) do not alter that risk.

Mainstream macroeconomics is mired in the flawed belief that the government has three sources of funding: (a) taxes; (b) debt-issuance; and (c) printing money. Students rote learn that the degree of expansion is in order of these options and that printing money is “so expansionary” because it also reduces interest rates (in their models) as well as directly impacts on demand that it is inflationary.

Not only do sovereign government not have to “fund” their spending, the government budget constraint (GBC) framework which is the pedagogical device that students learn this nonsense within is deeply flawed.

Students learn that the printing money option – called monetisation – involves the government running a deficit with no external debt issued to “pay for it”. What a nonsensical description of the options facing a sovereign government. When the government spends the recipient receives a credit (say a deposit at a bank) at the same time the government credits the reserve accounts that the particular commercial bank maintains with the central bank. There is nothing more to it than that.

So bank deposits and reserves rise by an equal amount and the recipient expresses the increase in their net financial position (wealth) in the form of an increased deposit at the bank.

The banks now have excess reserves because they can always rely on the central bank to provide them with reserves should they be short. The central bank may offer a support rate on these reserves (as Australia has done for year – 25 basis points below the target rate; or as the US has done since 2008).

If there is no support rate, then the overnight inter-bank interest rate will drop to zero as a result of the desire of banks to seek some overnight return on these reserves. The reality is that these transactions among banks net to zero which means they cannot eliminate a system-wide surplus. Assuming the central bank has a non-zero target rate then it has to “drain” these excess reserves (a liquidity management operation) to prevent the competition among banks to lend their excess reserves.

How does it do this? Answer: sell government bonds. What does this mean for the textbook version of “monetisation”? Answer: it means the mainstream textbook rendition is wrong. The central bank has no choice but to sell debt in this situation or else it has to be prepared to run a zero interest rate regime (as in Japan for two decades).

In the current period, where the central bank offers to target rate to banks holding overnight reserves, these liquidity management operations are no longer required. This means that the central bank can ignore the impacts of the deficit on bank reserves and maintain control over the interest rate by ensuring it pays the same rate on overnight excess reserves as it announces as its policy rate. There are no further complications with this approach.

The national government does not have to issue debt to net spend (that is, run a deficit). So wouldn’t this be more inflationary than if it issued debt? Answer: not at all. The inflation risk is in the spending not the monetary operation.

To understand this one has to come to terms with the stock impacts of the deficit flows. Government deficits increase non-government sector net savings by increasing national income. The stock manifestation of this net saving flow manifests as increased net financial assets (wealth) held by the non-government sector initially as bank deposits.

What if the government issues debt as described above? Answer: the portfolio of the net financial assets held by the banks is changed. The bank debits its reserve account (at the central bank) by the sum of the payment for the government bonds and it now holds paper instead of reserves on its balance sheet. The reserves are “drained” from the system by the monetary operation.

Have the deposits held by the initial recipient (which was the expression of the rise in net financial assets due to the deficit) disappeared? Answer: No, they are still there.

What if the government issues debt to the public? Answer: Bank reserves still decline because the holders of deposits at the bank pay for the bonds via a reduction in their deposit balances. The improved net financial asset position of the non-government is now expressed by the holding of the government debt by the public (not banks) rather than bank deposits. The public have just converted their non-interest bearing wealth into interest-bearing wealth.

Has there been any difference in the change in aggregate demand as a result of these different options? Answer: none, aggregate demand went up by the same amount equal to the net government spending plus the multiplied induced consumption. Please read my blog – Spending multipliers – for more discussion on this point.

So you can see the fallacy in the theories that distinguish between “printing money” and “issuing bonds” and conclude that one is more inflationary than the other. There is no difference for aggregate demand. To understand this you need to also appreciate that banks do not lend reserves.

Bank reserves are an essential part of the clearing or payments or settlements system (different terminology is used in different countries to denote the same mechanism). Accordingly, the funds are used to settle the myriad of transactions that implicate each of the banks each day. Having excess reserves doesn’t increase the capacity of the banks to lend.

In a fiat monetary system, the banks can lend whenever they choose. Their loans create deposits and are not dependent on the existence of “excess” reserves. Banks will lend to anyone who they can make a profit on. If they didn’t have the required reserves at the end of the day then the central bank provides them.

So there is no sense in the logic that holdings of excess reserves are inflationary. Bank reserves are not a component of aggregate demand.

Please read the following blogs – Building bank reserves will not expand credit and Building bank reserves is not inflationary – for further discussion on this topic.

What about the argument that if the non-government sector has bonds they are less likely to spend these “savings”? Answer: If the holders of either form of net wealth (bank deposits or bonds) wanted to dis-save they could easily do that. In this instance, the budget deficit would fall anyway as the automatic stabilisers increased tax revenue and welfare spending fell. If the automatic stabilisers didn’t reduce government demand enough – that is, nominal demand growth was outstripping the real capacity of the economy to produce more goods and services then further fine-tuning of discretionary fiscal policy would be required to avoid inflation.

The point is that public and private spending works in more or less the same way. Both add to demand and command real resources and both can be excessive in relation to each other and the real capacity of the economy to produce goods and services.

Please read Scott Fullwiler’s excellent coverage of this issue – HERE – for more detail in the US context.

So these deep anxieties about inflation and deficits are ill-founded and should not lead to the policy debate being derailed into sideshows like debates about the merits of a return to a Gold Standard.

Brief review – How did a Gold Standard work?

Please read my blog – Gold standard and fixed exchange rates – myths that still prevail – for an introduction to this topic.

I won’t repeat that blog and the nuances of pure commodity currency systems and convertible currency systems backed by gold are explained there.

Zoellick is thinking of a return to a paper currency system where paper money issued by a central bank is backed by gold. So the currency’s value can be expressed in terms of a specified unit of gold. To make this work there has to be convertibility which means that someone who possesses a paper dollar will be able to swap it (convert it) for the relevant amount of gold.

In the early versions of the Gold Standard, central banks would set their currency against the gold parity. As an example, say the Australian Pound was worth 30 grains of gold and the USD was worth 15 grains, then the 2 USDs would be required for every AUD in trading exchanges.

The central banks would maintain the “mint price” of gold fixed by buying or selling gold to meet any supply or demand imbalance as a result of trade between nations. Further, the central bank had to maintain stores of gold sufficient to back the circulating currency (at the agreed convertibility rate).

Gold was also considered to be the principle method of making international payments. Accordingly, as trade unfolded, imbalances in trade (imports and exports) arose and this necessitated that gold be transferred between nations (in boats) to fund these imbalances. Trade deficit countries had to ship gold to trade surplus countries.

So trade surplus nations could expand their domestic money supplies (issue more notes) because they had more gold to back the currency. This expansion was in strict proportion to the set value of the local currency in terms of grains of gold. The rising money supply would push against the inflation barrier (given no increase in the real capacity of the economy) which would ultimately render exports less attractive to foreigners and the external deficit would decline.

For the deficit nations, the loss of gold reserves forced their governments to withdraw paper currency which was deflationary – leading to rising unemployment and falling output and prices. The latter improved the competitiveness of their economy which also helped resolve the trade imbalance. But it remains that the deficit nations were forced to bear rising unemployment and vice versa as the trade imbalances resolved.

The political costs of this imbalance in domestic fortunes ultimately led to the system being abandoned (in an evolved form) in 1971.

The proponents of the gold standard focus on the way it prevents the government from issuing paper currency as a means of stimulating their economies. Under the gold standard, the government could not expand base money if the economy was in trade deficit. It was considered that the gold standard acted as a means to control the money supply and generate price levels in different trading countries which were consistent with trade balance. The domestic economy however was forced to make the adjustments to the trade imbalances.

Monetary policy became captive to the amount of gold that a country possessed (principally derived from trade). Variations in the gold production levels also influenced the price levels of countries.

In practical terms, the adjustments to trade that were necessary to resolve imbalances were slow. In the meantime, deficit nations had to endure domestic recessions and entrenched unemployment. So a gold standard introduces a recessionary bias to economies with the burden always falling on countries with weaker currencies (typically as a consequence of trade deficits). This inflexibility prevented governments from introducing policies that generated the best outcomes for their domestic economies (high employment).

Ultimately the monetary authority would not be able to resist the demands of the population for higher employment.

After World War 2, the IMF was created to supersede the gold standard and the so-called gold exchange standard emerged. It is this sort of system that Zoellick is proposing.

Convertibility to gold was abandoned and replaced by convertibility into the USD (although Zoellick would like a broader basket of reserve currencies), reflecting the dominance of the US in world trade (and the fact that they won the war!). This new system was built on the agreement that the US government would convert a USD into gold at $USD35 per ounce of gold. This provided the nominal anchor for the exchange rate system.

The Bretton Woods System was introduced in 1946 and created the fixed exchange rates system. Governments could now sell gold to the United States treasury at the price of $USD35 per ounce. So now a country would build up USD reserves and if they were running a trade deficit they could swap their own currency for USD (drawing from their reserves) and then for their own currency and stimulate the economy (to increase imports and reduce the trade deficit).

The fixed exchange rate system however restricted fiscal policy (which goldies consider an advantage) because monetary policy had to target the exchange parity. If the exchange rate was under attack (perhaps because of a balance of payments deficit) which would manifest as an excess supply of the currency in the foreign exchange markets, then the central bank had to intervene and buy up the local currency with its reserves of foreign currency (principally $USDs).

This meant that the domestic economy would contract (as the money supply fell) and unemployment would rise. Further, the stock of $USD reserves held by any particular bank was finite and so countries with weak trading positions were always subject to a recessionary bias in order to defend the agreed exchange parities. The system was politically difficult to maintain because of the social instability arising from unemployment.

So if fiscal policy was used too aggressively to reduce unemployment, it would invoke a monetary contraction to defend the exchange rate as imports rose in response to the rising national income levels engendered by the fiscal expansion. Ultimately, the primacy of monetary policy ruled because countries were bound by the Bretton Woods agreement to maintain the exchange rate parities. They could revalue or devalue (once off realignments) but this was frowned upon and not common.

Whichever system we want to talk off – pure gold standard or USD-convertible system backed by gold – the constraints on government were obvious.

The gold standard as applied domestically meant that existing gold reserves controlled the domestic money supply. Given gold was in finite supply (and no new discoveries had been made for years), it was considered to provide a stable monetary system. But when the supply of gold changed (a new field discovered) then this would create inflation.

So gold reserves restricted the expansion of bank reserves and the supply of high powered money (government currency). The central bank thus could not expand their liabilities beyond their gold reserves (although it is a bit more complex than that). In operational terms this means that once the threshold was reached, then the monetary authority could not buy any government debt or provide loans to its member banks.
As a consequence, bank reserves were limited and if the public wanted to hold more currency then the reserves would contract. This state defined the money supply threshold.

The concept of (and the term) monetisation (as discussed above) comes from this period. When the government acquired new gold (say by purchasing some from a gold mining firm) they could create new money. The process was that the government would order some gold and sign a cheque for the delivery. This cheque is deposited by the miner in their bank. The bank then would exchange this cheque with the central bank in return for added reserves. The central bank then accounts for this by reducing the government account at the bank. So the government’s loss is the commercial banks reserve gain.

The other implication of this system is that the national government can only increase the money supply by acquiring more gold. Any other expenditure that the government makes would have to be “financed” by taxation or by debt issuance. The government cannot just credit a commercial bank account under this system to expand its net spending independent of its source of finance. As a consequence, whenever the government spent it would require offsetting revenue in the form of taxes or borrowed funds.

Ultimately, Bretton Woods collapsed in 1971. It was under pressure in the 1960s with a series of “competitive devaluations” by the UK and other countries who were facing chronically high unemployment due to persistent trading problems. Ultimately, the system collapsed because Nixon’s prosecution of the Vietnam war forced him to suspend USD convertibility to allow him to net spend more. This was the final break in the links between a commodity that had intrinsic value and the nominal currencies. From this point in, most governments used fiat currency as the basis of the monetary system.

So would a return to a Gold Standard alter any of this?

Answer: No! We would return to the same problems that led to the system’s downfall. In a sense, the Eurozone is running a version of the Gold Standard – in the form of a fixed exchange rate system (with no prospect even of re- or devaluation). The only way that a Eurozone nation can adjust to a persistent external deficit is to implement a harsh domestic deflation to increase competitiveness. It is a path to falling living standards and ultimately political and social instability.

There are claims that the international currency system is in a state of near collapse.

The UK Guardian economics correspondent Larry Elliott wrote (November 8, 2010) that:

… without doubt, a return to gold has some attractions. The international currency system is in turmoil; countries are adopting beggar-my-neighbour devaluations; the gold price has soared to just shy of $1,400 (£868) an ounce and this week’s G20 meeting in South Korea is shaping up to be an ill-tempered affair.

The devaluations (or managed rates) are also manifestations of the destructive effects of the neo-liberal era exemplified by the IMF and World Bank obsession with export-led growth. The underpinning agenda of this approach can be traced back to a anti-fiscal policy obsession. Export-led growth strategies are not reliable and deprive the citizens of access to the fiscal capacity of their government and the use of their own resources.

Please read my blog – Export-led growth strategies will fail – for more discussion on this point.

Each government can make its own decision about whether they choose to hold USD reserves or diversify, say into Euro. But they would be far better advised to stop seeking to net export as a policy position and instead use their fiscal capacity to stimulate domestic demand. That would solve this “alleged” international currency system turmoil pretty quickly.

I also find it amazing that nations like China and Germany are now lecturing the Americans for “living beyond their means” and demanding the US government cuts back on US demand. If the US did cut back on demand for imports then where are these export goliaths going to make their money from? Answer: this posturing is total hypocrisy.

Elliot does make there good reasons why a return to the gold standard would be disastrous. He said:

Firstly, the world supply of gold is too small to support a global economy with an annual output of $50tn a year. Secondly, the global economy is a lot more diverse than it was in the 19th century, when the scene was dominated by a handful of European nations plus America. Thirdly, pegging currencies to gold would almost certainly prove to be deflationary. Here, the lesson of Britain is apposite, since the 1925 decision by Churchill to return to the gold standard at the pre-war parity following pressure from the Bank of England governor Montagu Norman was one of the great economic blunders of the 20th century. Within six years, deflationary pressure had forced Britain to abandon gold.

All those sovereign nations running current account deficits take note! If they support a return to a gold standard at the current G-20 meeting then they are also signalling they support government policy deliberately creating unemployment and poverty and deflation.

The problem is that the proponents of these mad schemes don’t make that connection. If it was their job at stake maybe they would.

For the Euro nations – they should abandon their monetary union and by floating their reintroduced currencies start improving the living standards of their citizens and eschewing the destructive path they are on at present. Either way the path forward is now going to hurt these nations but in the medium- to long-term these nations will be far better off abandoning the Euro, placing a moratorium on debt, protecting their housing markets and getting people back into work and spending locally.


To move forward the world leaders have to abandon all these neo-liberals manias and see that spending equals income. In normal times, public spending can be moderated by private spending growth to ensure that inflation is contained and employment growth is high. But these are not normal times and there is little to suggest that private demand growth in most of the advanced nations is going to become very robust anytime soon.

In that context, given spending equals income, the only show in town is to expand fiscal policy. By sidetracking into these non-issues – quantitative easing; export-led growth; gold standards etc the leaders are just going to make things worse.

I liked Brad Delong’s conclusion – Zoellick “really may be the Stupidest Man Alive”.

That is enough for today!

Spread the word ...
    This Post Has 151 Comments
    1. Vimothy,

      Do not disagree with you on the description.

      Firstly there is correlation. Of course there will be – over time prices rise and money supply also rises. Its trivial.

      Secondly, you can see that money is not neutral.

      Thirdly, not only is correlation not causation, correlation is not beta. ρ≠β. To exaggerate a little, I can have two correlated variables – with high correlation but one in which if one jumps by 100% the other jumps only by 1%. Clearly these two are highly correlated by construction but you see if one jumps by 200%, the other one jumps only 2%.

      I see you know the numbers well, but I do not see why you think that if the money supply increases by 10%, prices will also rise by 10% – the charts in Mervyn King’s paper do suggest that its not the case.

      Also he himself suggests that correlation is not causation!

      So going back, I never claimed there is no correlation. I am claiming that the causality is the other way round. More importantly, money is not neutral but the money stock is a meaningless abstraction – but thats a topic for another day.

    2. Tom,

      I’m having trouble seeing your point. Change in the quantity of money with respect to inflation is indeed meaningful. In the long run, one seems to determine the other–or something like that. Change in the quantity of M alone, is just change in the quantity of M. It’s meaningful if we’re interested in the change in the quantity of M.

      What is the basis for your second paragraph? Most economists do not focus on the nominal at the expense of the real, in my experience. I’m not really sure what you mean here.

    3. “So going back, I never claimed there is no correlation. I am claiming that the causality is the other way round. More importantly, money is not neutral but the money stock is a meaningless abstraction – but thats a topic for another day”

      Okay thanks for clarifying. At first you seemed to be suggesting that but then in response to me it looked like you were denying the existence of the correlation. Glad we’re on the same page. So my original question was:

      Given the long-run co-movement between the two variables, your claim is that inflation causes money growth (unlike the QTM story where the causation runs in the opposite direction). What then causes inflation in your opinion?

      And is their correlation really trivial? I can think of lots of ramifications, e.g. one extension might be that in the long run increasing agg demand doesn’t increase output.

      If we wanted to estimate a simple linear regression of log(inflation) on log(moneygrowth), for example, then our population model would be of the form log(inflation) = β_0 + β_1log(moneygrowth) + u. A one-to-one relationship implies that β_1 = 1. It might or it might not, but (from the graphs) clearly the population parameter is greater than zero and approaching 1 in the long run. Remember though that we are talking about the long-run. In the short-run, the relationship between these two variables is a lot less clear. In the long-run, we would predict that if money supply increased by 1%, expected inflation growth would increase by β_1%.

      King suggests that both the variables are endogenous, which seems to me to be the most sensible position. Which is why I’m interested in finding out why you disagree.

    4. Vimothy,

      I will concentrate on King’s paper because its all there.

      The β_1 is not really 1 :) in either the short run or the long run but one may say that it looks as if it is trending to 1. One can interpret it that way and it is difficult to present a solid argument that its not to be honest. So what I am saying is that one can come up with a fit for β_1 versus time horizon in which is approaches one when the time horizon is taken to a large value and another fit using a different function where it does not and comfortably misses hitting 1.

      The fact that that’s the case means that its not easy to resolve the differences. Its good however that Mervyn King is open about this and is not really behaving like a Monetarist. My earlier view of the paper was my own interpretation of how you interpreted it.

      So it gets down to how money is created which is where you are aiming to head. And what causes inflation. Now I didn’t really mean that inflation causes money – what I meant was that it is one of the factors. Money (as in deposits) is created when a loan is made. It is destroyed when a loan is repaid. There are other things as well – interest paid on bank loans reduces money supply whereas interest paid by banks on deposits increase the supply. Plus there are even more factors. When governments spend, it increases the supply and when taxes are paid or government securities are purchases by the non-bank sector, deposits go down. There are of course other things which change the deposit levels and you have to look into it in detail.

      So the irrelavance of money can be seen by the following – if producers use the bank for borrowing, it increases the money supply but if they use the financial markets through issuance of debt securities, it doesn’t increase deposits – the deposits just change hands. So for believers that money causes inflation, the question that can be posed is – does borrowing through bonds cause less inflation than borrowing from banks ?

      The central mistake of Monetarists is that if demand increases they think prices rise. Michal Kalecki’s observation was that it doesn’t – production i.e. quantity increases. There are different sectors in the economy such as the agricultural sector, manufacturing sector, services sector and one has to think how they respond to increased demand. If one concentrates on the manufacturing sector, one can think of wages, capacity utilization and factors such as that. Inflation may also happen due to supply shocks such as a food shortage etc. But these are supply facors. They are not due to demand directly. For example if the economy is away from full employment, workers will have low bargaining powers and wage increases may be less than the productivity increase. So one factor to keep in mind is wage rise compared to productivity.

      More generally you have to look at the relationship between banks and their customers and banks and the central bank. Do banks wait for the central banks to create reserves and then lend or is it the opposite ? Such questions.

    5. Vimothy: I’m having trouble seeing your point. Change in the quantity of money with respect to inflation is indeed meaningful. In the long run, one seems to determine the other–or something like that. Change in the quantity of M alone, is just change in the quantity of M. It’s meaningful if we’re interested in the change in the quantity of M.

      I am saying that the change in q is economically relevant only in relation to changes in other — real — factors. Then correlation becomes interesting, but the principal question remains establishing causation.

      What is the basis for your second paragraph? Most economists do not focus on the nominal at the expense of the real, in my experience. I’m not really sure what you mean here.

      The great debate in the US now is not about unemployment (real), which is what polls show the majority of the country is chiefly concerned with, but about supposedly deficit and debt levels (nominal) that “everyone knows” are excessive. It’s beyond stupid. It is destructive.

    6. Ramanan,

      A few points:

      We don’t observe population parameters so it isn’t possible to say with certainty what value β_1 takes. The best we can do is estimate.

      You don’t need to be a monetarist to believe in the long-run neutrality of money.

      If demand increases, prices might increase or they might not. You can’t say a priori with no other knowledge that they will not increase. If there is no spare capacity in the industry receiving the extra spending, and production cannot increase, then prices must.

    7. Vimothy,

      Its a possible scenario that your β_1 approaches 1 in the long run and plenty of evidence that it doesn’t – our standards of living have improved a lot!

      The important thing to note – which you may have missed in my comment is to understand the money creation process to appreciate the irrelevance of the stock of money. Its a residual.

    8. Tom,

      I think you mean change in P. A change in Q would be what you describe as a “real factor”. The (posited) LR neutrality of money is interesting if you are intersted in macroeconomics. Otherwise I could see how you would find it to be irrelevant. I haven’t met many girls who are impressed by my ability to explain BOP imbalances or open market operations either. Maybe I just move with the wrong crowd.

      That said, I agree that there is too much said about govt debt and not enough said about unemployment. (And I’m don’t know if you’re aware, but you are using the terms “real” and “nominal” in ways that are somewhat unusual).

    9. Ramanan “Its a possible scenario that your β_1 approaches 1 in the long run and plenty of evidence that it doesn’t – our standards of living have improved a lot!”-

      -Is it possible that currency expansion has two effects. It both creates inflation (over the long term) but also increases the total purchasing ability of the nation with the expanding currency in relation to the rest of the world. That second effect could account for the increase in US standards of living over the last few decades and also the pronounced decrease in the standard of living of the least well off 25% of the world’s population over that time frame.

    10. Ramanan,

      “Its a possible scenario that your β_1 approaches 1 in the long run and plenty of evidence that it doesn’t – our standards of living have improved a lot! ”

      What evidence?

    11. vimothy, “It would violate logical consistency if money was both neutral and non-neutral in the long-run.”-

      -I guess I was meaning that what prevents expansion of say total USD from being neutral in the long run is the way that USD expansion leads to transfer of purchasing ability from the rest of the world to the overall USD stock (so even though any single $ purchases less the combined total USD can purchase more).

    12. vimothy “But monetary expansion IS neutral in the long-run, so finding hypothetical explanations for LR non-neutrality doesn’t make sense.”-

      Since monetary expansion is an ongoing process we are all always living in the short term not the long term. In the short term, currency expansion causes much more pronounced asset price inflation than consumer price inflation and so the effect is very much non-neutral in terms of the effects on wealth distribution. It makes the rich richer at the expense of the poor -both on a national and global scale.

    13. Vimothy,

      The evidence is right there in Mervyn King’s paper – am not sure why you don’t see it! The fact that the points are so much below the 45 degree line even for the 30-year graph. Even King doesn’t go as far to make a conclusion that money is neutral. For points which are above the 45 degree line – how about looking it this way – annual inflation was 5% but money supply didn’t expand 5% ???

      Now the point is one can keep arguing about this but its like arguing but only if you actually go into the details on how money is created, can you understand whats going on. Else its all difficult to explain to an Aristotelean that “objects in rest or in motion continue to be in rest or in motion …”

      Sorry couldn’t catch your point about comment on relation with the rest of the world.

    14. Ramanan,

      Those charts are far from being evidence of LR monetary non-neutrality. If money was non-neutral in the long-run then the datapoints wouldn’t have arranged themselves so obligingly on the 45 degree line. Of course, the points do not lie exactly on the line. Even if it were a fitted line from a linear regression we wouldn’t expect all the datapoints to lie on it. Hence the error term. But, roughly, what is the relationship? (Or, if you prefer, what is the “expectation” of inflation conditional on money growth?) It’s clear that in the long-run, inflation growth and monetary growth move together, while there does not appear to be a relationship between monetary growth and GDP growth.

      The details of money creation are important (and very interesting), but I’m not sure how relevant they are here. Happy to be persuadeed otherwise, of course.

    15. vimothy: (And I’m don’t know if you’re aware, but you are using the terms “real” and “nominal” in ways that are somewhat unusual).

      Sorry for the confusion. I was using “nominal” and “real” in their philosophical sense (I am a philosopher rather than an economist). I did not mean not-inflation adjusted v inflation-adjusted but rather “notional” v. “actual.” My point is that a lot of contemporary economics seems to be fixated on the notional so that the debate becomes disconnected from the actual.

      When the debate becomes one of changes in the monetary base representing quantity of money (as now), the debate becomes disconnected from reality, because the size of the monetary base does not affect supply and demand directly (as erroneously presumed). Nor is inflation correlated with deficits adding to nongovernment NFA. The US has been running large deficits with low inflation. Nor is inflation correlated with M1 (chart). However, it can be argued that inflation is correlated with the exchange rate and the price of petroleum (chart).

      So inflation is clearly not directly correlated to changes in the monetary base, as QE1 shows. Nor is inflation is correlated with changes in endogenous money supply (M1, M2, M3). See Ben Bernanke, Monetary Aggregates and Monetary Policy at the Federal Reserve: A Historical Perspective

      But since the endogenous money supply is dominated by credit money, the process of credit extension is asset-inflationary, i.e., prices tend to exceed value, leading to financial instability, as Minsky’s work shows.

    16. Dear vimothy (at 2010/11/20 at 2:13)

      You keep asserting the Classical/Monetarist religious diatribe:

      But monetary expansion IS neutral in the long-run

      Kalecki said that the economy was just a sequence of “short-runs”. What exactly is the long-run? Where is the evidence and theoretical case (that is logically consistent) to say that monetary expansion has no real effects on an on-going basis? Do you reject path-dependency (yes!)? On what basis (none!)? How are you measuring money? Do you think it is somehow exogenous to the model (you must!)? On what grounds are you rejecting the endogenous growth theories of money that show that what is “measured” as m3 or whatever is a meaningless artefact of the credit creation process?

      And a lot more.

      The classical dichotomy which you are continually asserting here (money neutrality) was shown to be logically inconsistent years ago.

      best wishes

    17. Bill,

      I mentioned upthread that in my view, money and inflation are both endogenous variables. I guess this is unfashionably mainstream round here, but there it is. Kalecki’s epithet is pithy, but I don’t find it to be a particularly substantive critique. “The long-run” is a just a bunch of words. It doesn’t have to take on a single and exact constant value. In which case, quoth Kalecki, it cannot exist…? Between him and my own lying eyes, it may be a struggle, but it’s not much of an argument. “It depends”.

      When I look at the charts in King’s paper, I notice that the the distribution of datapoints in the 30 year scatter plot is different to the 1 year scatter plot. Over the 30 year horizon, the data groups tightly about the 45 degree diagonal. Over the 1 year horizon, this is not the case. There is no correlation in the money vs output scatter plots at any length of time. Benati’s paper uses a technique for filtering the components of time series oscillating at different frequencies. But it’s basically the same insight: change the resolution to get a better look at the picture. I find that, looking at the charts in these papers, in the long-run (sorry!), money and inflation move together. On the other hand, I don’t see any relationship between money and GDP.

      I won’t go over the theoretical case, since it is so straightforward and widely known. It strikes me as the most parsimonious and intuitively correct answer–if the price level changes, nothing has really changed. There are lots of papers cited in the two QB articles for anyone looking to take this further.

      Measuring money: again, like the piece of string, it depends. But the relationship seems to be there whatever measure you pick (see the charts). Of course, these assets exist on a spectrum, but the “moneyness” of any asset must surely matter a great deal if we want to consider its contribution to inflation of whatever type.

      I have no stake in this, by the way. It is a matter of total indifference to me. But I don’t see any other way to interpret the evidence.



    18. Tom,

      We have similar interests! I studied philosophy during my undergrad degree before switching to English lit. Good times.

      However, you’re missing my point. I’ve already linked to some charts that show the correlation between money and inflation. (King’s paper, “no money, no inflation”, previous page, charts 1-). Have you seen them?

    19. I have many charts that show that there is strong correlation between housing and population. Clearly, if we were to increase the stock of houses, then population would also increase. At least, over the long run.

      Therefore I’m worried that if we build too many houses, that this will give rise to overpopulation.

      We have to really be careful here, and make sure that there is some constant level of homelessness, because if we were to supply everyone with a house, then the population would explode.

    20. Oh, for god’s sake. There are so many straw men being wrestled here that I’m beginning to think lighting this cigarette was a bad idea.

    21. Well, Vimothy, what do you expect? Your whole analysis is a stock/flow error.

      You are confusing money with income and have no micro-basis for a belief that an increase in the stock of base money will result in an increase in income flows. The stock of base money circulates and can support large variations in income flows.

      But clearly if income flows continue to increase, then at some point the stock of base money becomes a constraining factor, and money demand by the banks goes up, so that overnight rates go up, at which point the CB will create more money to meet that increased money demand if it wants to keep the rate at target.

      So one drives the other but not the other way around. You can cite correlations until you are blue in the face and it wont change the fact that your micro analysis is all wrong.

    22. Instead appearing out of nowhere and throwing in game changers like “correlation is not causation”, why don’t you read my posts and respond to them?

    23. Perhaps it’s just me, but at times it feels like its impossible to have rational debate about an issue without being drawn inexorably towards the MMT catechism…


      “Yeah, I know that, but what does that have to do with…”


      “I agree, but…”


      “Er, yeah, right, but…”



      Like trying to make small talk with a dallek. Deep breaths now.

    24. Vimothy,

      You said

      Measuring money: again, like the piece of string, it depends. But the relationship seems to be there whatever measure you pick (see the charts). Of course, these assets exist on a spectrum, but the “moneyness” of any asset must surely matter a great deal if we want to consider its contribution to inflation of whatever type.

      I have no stake in this, by the way. It is a matter of total indifference to me. But I don’t see any other way to interpret the evidence.

      And I was providing an interpretation for the relationship.

    25. Vimothy,

      You are making some assumptions here. If I approach toward a stationary object, it doesn’t mean I reach the object. My speed may get progressively slower and there may be an a finite distance even if I live forever.

      You are interpreting the graphs in King’s paper as this: for 1y/2y its far from the 45 degree line. For 5y the points seem to closer to the 45 degree line than it was. Similarly for 10y and hence you are saying that the points will strongly be attracted to the 45 degree line if I take a sufficiently long time period not 30y. Need not be the case – just the the analogy I gave you.

    26. .. and thats reverse Zeno’s paradox.

      Zeno said that to cross a ground, I need to cross half of it. After I cross half of it, I need to cross half of the remaining half .. ad infinitum … and hence I will never be able to cross the ground.

    27. Nevertheless, as t approaches 30, the data clusters ever more tightly about the y=x line. I find this significant, but mileage, to paraphrase Mencius Moldbug, is also endogenous–so why should you? No reason at all, I guess.

      In my view both variables are endogenous. It doesn’t make sense to say that movement in one caused movement in the other. Do you disagree? I can’t tell. Maybe I disagree–maybe I do think that money causes inflation. I’m starting to lose the thread here. I think I could do with a stiff drink, so with that in mind…

    28. vimothy, of course there is “inflation in the long run” because cb’s target about 2% inflation a year in order to incentivize investment and consumption over saving. It’s part of the design. Does this prove that increasing quantity of money “causes” inflation? It would seem not since the cb targets its desired inflation rate using interest rate adjustment rather than quantity.

    29. Vimothy,

      Nevertheless, as t approaches 30, the data clusters ever more tightly about the y=x line.

      That’s how you look at it. Would have believed you if points were spread on both sides of the 45 degree line. Equally and tightly. There is a definitive bias to values being below the 45 reason line and hence the stylized fact “money is not neutral”

      Unfortunately there is no data – just graph. Some points mean money supply rose 17% in one year but annual inflation was 4% or so.

      Here is what you should do – get the data for 119 countries and take some weighted average mean of y/x and show that it is sufficiently close to 1.

      And its not the case with any of the big economies (the data I have actually checked) – you can go to their national statistics websites, get the data on an excel sheet and convince yourself.

      Perhaps the bigger point gets missed in debates such as this. The definition of money itself. Different countries have measures such as M0, M1, M2, M3 and even M4! Each has its own definition of what is to be included in which category. Does one include money market mutual funds shares ? Because its equivalent to bank deposits in most respects.

      I am not sure why we are debating this because it looks similar to two people looking at two stars and disagreeing on which is further.

    30. Ramanan:

      “Does one include money market mutual funds shares ? Because its equivalent to bank deposits in most respects.

      Inasmuch as one considers one’s house, or one’s Microsoft shares, or one’s bonds deposits.

    31. Tom, can you at least see why I’m confused? I’ve said three times that money does not “cause” inflation–both variables are endogenous. Money does not cause inflation, inflation does not cause money. However, in the long-run, money growth and inflation move together.

    32. Tom,

      BTW, you seem to be arguing in some posts that money and inflatoin do not move together (eg at November 20, 2010 at 5:26) and in other posts that this correlation is obvious: “of course there is “inflation in the long run””.

    33. VJK,

      Yes I know.

      You seem to have gotten into discussion about this a lot here or some other place.

      Doesn’t matter what one’s definition is – one spends out of income, not deposits. Spenders make two decisions – consumption and allocation of the saving. There is a hierarchy of decisions. So “spending out of deposits versus spending out of bonds” is meaningless.

    34. Ramanan,

      I agree with you, but this is not a calculus class. In the limit, as n increases, the accuracy of our parameter estimate should also increase. As I said, its value is closer to 1 than 0. But we don’t need to be precise, just understand the relationship.

    35. Last word for now:

      RSJ, you are normally so insightful, I can’t believe you have nothing of value to add beyond clichés. To save you the trouble of having to read my previous posts, let me restate my position: in the long run, money and inflation move together. This relationship is relatively stable and holds across different policy regimes. In the LR, inflation is monetary.

    36. Ramanan:

      “Fed’s M3, which it no longer publishes included money market mutual funds shares”

      But that was quite daft, wasn’t it ? “Other liquid assets” is really funny. The Bernank got at least *that* right, having got rid of M3.

      Why not include Microsoft’s shares for good measure, or Berkshire Hathaway’s ?

      “Any tradable IOU is money”, right ?

      As Vimothy put it “I have no stake in this, by the way”. Just being a curious and humble practitioner of the “art” rater than an economist.

    37. BTW, you seem to be arguing in some posts that money and inflatoin do not move together (eg at November 20, 2010 at 5:26) and in other posts that this correlation is obvious: “of course there is “inflation in the long run””.

      This is the fundamental issue. Those putting forward the quantity theory concluded that because of correlation “in the long run” that there must be some nexus that could be used to predict and control in the short run. But astute traders gave up on the predictability of the quantity theory some time ago, and the Fed found it useless for control, too. When one tries to pin down the supposed connection, it is elusive because it is not there. This is sort of like the Samuelsonian myth of “the invisible hand” as the basis for general equilibrium. See Gavin Kennedy, General Equilibrium and the Myth of Invisible Hand

      If you want to claim that over time money stock and inflation grow with a growing economy, no one will argue with that, however. A growing economy with increasing population and increasing productivity can’t grow without increasing funds to support increased transactions, and the Fed builds in some inflation for velocity. So no surprises there. But it’s rather trivial and doesn’t say much.

    38. Dear Tom Hickey (at 2010/11/21 at 9:42) and all in this debate

      The assertion that there is a “long-run” relationship that is proportional between some measure of money and some measure of inflation requires an analytical and empirically-tractable definition of the long-run. The King paper does not provide that definition. Why is 30 years the long-run? In economics it means some special (capital variable and expectations fully adjusted) and of-course the conditions are never met. That is why Kalecki said that the world was just a sequence of short-runs.

      Further, to think that the proportionality between nominal aggregates is interesting you have to have a reason. For the classical/new classical mainstream it was a way of maintaining their assertion that there was full employment (or natural rate of unemployment) in the “long-run”. We rarely observe anything like full employment and there is often excess capacity. So even in the new Keynesian models which assert neutrality in the “long-run” (for them it is when expectations are adjusted) there is always non-neutrality because there is typically excess capacity.

      It goes without saying that if you have full capacity operation and you expand nominal demand then you get inflation. That is not a very interesting observation.

      Finally, the very fact that one considers monetary growth to be endogenous raises the question: what motivates that endogeneity? Credit expansion in an endogenous money world is about firms borrowing to buy working capital to produce real things. So if you believe that monetary growth is endogenous you will then have a hard time saying that it is neutral.

      best wishes

    39. I wasn’t actually thinking of King, but of Beryl Sprinkel. Back in the ’60’s and early ’70’s he was the rage among traders. We all followed his Harris Bank letter and read his books, Money and Stock Prices, and Money and Stock Prices. He was a monetarist that other will probably remember as President Reagan’s chairman of the Council of Economic Advisors.

      Anyway, he was making a case of the connection of changes in interest rates, money supply, and stock prices, and many traders, including me, thought that he was really on to something. The data and argument all looked good, and the Fed, too, was taking a monetarist approach. I don’t know of any traders who made it work, although some are still trying, and the Fed subsequently abandoned targeting money supply, too. The data seems convincing, but no one ever got the theory to actually work. My point to vimothy was that what may seem “obvious” from looking at data ain’t necessarily so when the rubber hits the road in trading or conducting policy. Been there, done that.

      Anyway, those interested in the history might want to take a look at this research paper of the FRB of St. Louis that summarizes the thinking that was going on then. Warning: very wonkish. But just perusing it, you get the idea. Impressive math and essentially useless in the real world.

      Expectations, Money, and the Stock Market* by MICHAEL W. KERAN (1971)

      Conclusion: The intent of this article is threefold. First, it seeks a rational explanation for movements in stock prices which is consistent with standard economic price theory, and which can be tested against historical observations. It is shown that the standard theory of stock price determination, that is, discounting to present value expected future earnings, provides a solid theoretical base for a reasonably good empirical explanation of stock price movements in the past fifteen years. The major factors determining stock prices are shown to be expected corporate earnings and current interest rates. The interest rate in turn is determined by expectations of inflation, the real growth rate, and the change in real money. Increased earnings expectations tend to increase the stock price, while increased interest rates tend to depress the stock price. According to this analysis, changes in the nominal money stock have little direct impact on the stock price, but a major indirect influence on stock prices through their effect on inflation and corporate earnings expectations.

      The second objective of this article is to test the interrelationships between the stock price hypothesis and a monetarist econometric model of the United States. By integrating the stock price snbmodel into the monetarist model to obtain a combined model, it is possible to better understand the link between Federal Reserve actions (measured by changes in the nominal money supply) and the resulting effect on the stock and bond markets.

      A final objective is to illustrate how a small monetarist econometric model can be used to analyze subsectors of the economy. In this regard, the article can be viewed as an application of a monetarist model to issues with which the model was not originally intended to deal. The fact that it has worked with relative success provides further evidence on the usefulness of the monetarist model and its potential for further application in explaining other subsectors of the economy.

    40. Ramanan:

      A natural question would be “what MMMF” ?

      One could sell his Microsoft share in 2008 in a second.

      In 2008, one would not be able sell his commercial paper MMMF share without breaking the buck unless the Governement came for rescue.

      Remember those times ?

      In any case, why the Treasury bond is not money and a piece of commercial paper is money ? Is the former less liquid than the latter ?
      Obviously not.

      You have to be consistent in your logic ;)

    41. VJK,

      Not sure of your logic here. I do not pay close attention to what is counted as money and what is not.

      All economic concepts are raffish. However it won’t help comparing a money market mutual fund with a Microsoft share.

      Yes, the MMMFs were in trouble in 2008, but banks’ customer in many nations without deposit insurance also get into trouble.

      So would you classify such deposits as “not money” ?

    42. Or for that matter deposits upto $100,000 are insured in the US.

      Would you then not count deposits above $100,000 for customers as “not money” ?

    43. VJK,

      In any case, why the Treasury bond is not money and a piece of commercial paper is money ? Is the former less liquid than the latter ?
      Obviously not.

      I think you didn’t quite get me. MMMF shares offer transaction deposits like qualities and hence classified as near money or such. Now, you may not like the classification but that is your opinion.

      My definition was not just liquidity but liquidity and deposits-like quality.

    Leave a Reply

    Your email address will not be published. Required fields are marked *

    This site uses Akismet to reduce spam. Learn how your comment data is processed.

    Back To Top