Monday, August 22, 2011

Monetarism v. Fiscalism

MV = PY is an identity that means the amount of money (M) multiplied by the turnover of money (V for velocity) is identical to the amount of economic activity (Y) multiplied by the price level (P). MV = PY also appears as MV = PT. Here the T in place of Y stands for transactions.

What this says, in effect, is that purchases during a period are equal to sales over the period. Or, since what is spent is someone else's income, income equals expenditure over the period. MV represents money spent, and PY, or PT represents money earned. Give and take are always equal as a matter of accounting.

Generally, what is of interest in macroeconomics is Y, since Y is aggregate income (demand) and also aggregate product (supply), that is, Y (national income) equals GDP (gross domestic product). This is written as the identity, Y = C (for household consumption expenditure) + I (for firm capital expenditure) + G (for government fiscal expenditure) + (X - M), signifying net exports, that is, the trade balance. Summing this, national income (aggregate demand) equals national expenditure (aggregate supply).

Monetarism v. Fiscalism

On one hand, monetarists target money supply as the independent variable of economic policy since they view it as the chief means to control firm investment, hence, income leading to demand. For in their view, supply, which comes from investment, creates its own demand, which comes from income, iaw Say's law, in that firm expenditure funds household income.

On the other hand, fiscalists target income since income is the basis of effective demand, and in their analysis demand draws forth supply, since firms invest in response to effective demand for their goods. Effective demand sends a signal to firms to invest.

Strict monetarists reject fiscalism as ineffective. Fiscalists reject monetarism in a non-convertible floating rate system as inefficient if not also ineffective. New Keynesians accept fiscalism in a so-called liquidity trap but not otherwise. This is the basic kerfuffle going on in many current debates.

The basis of the debate between monetarists and fiscalists that is now raging is over whether monetary or fiscal policy is most appropriate, although some economists argue that neither is appropriate and the government should but out and let "the free market" take its course. This is crucial to understanding the debate between MMT and mainstream economists, most of whom are monetarists or have monetarist tendencies.


Monetarists are most concerned with M rather than Y, since their focus is inflation. While all economists agree that MV = PY, Milton Friedman used this identity to ground the quantity theory of money (QTM), which explains inflation in terms of change in M, or money supply. QTM holds that increases in M (money supply) imply a corresponding increase in P (price level), i.e., inflation, presuming that V (money turnover or velocity) and Y (product) are constant.

For monetarists, M (money supply) is the independent variable in MV = PY, changes in which influence the price level. So, according to monetarists, M needs to be controlled through changes in interest rates, since it is the interest rate channel that affects the relationship of saving and investment as primary determinants of economic activity.

The relationship of saving and investment is adjusted through interest rates to encourage investment without provoking inflation. In this view, saving funds investment. The central bank sets interest rates iaw inflationary expectations to adjust the balance of savings and investment, increasing rates to encourage saving when inflationary expectations rise, and lowering them to encourage investment when inflationary expectations are low enough.

This view presumes a credit-based monetary system, in which money is borrowed into existence from a central bank that is independent of the Treasury, or from the private sector, instead of being issued directly by the Treasury. This assumes that loanable funds are based on fractional reserve banking and the so-called money multiplier, such that deposits (saving) create reserves that fund loans. By adjusting reserves through monetary operations involving interest rate setting and reserve requirements, the central bank can therefore control the endogenous money supply in this view, and since the Treasury issues debt instruments sold to the private sector in order to obtain reserves needed for fiscal expenditure, it competes for loanable funds to the degree that expenditure exceeds revenue from taxation, thereby "crowding out" private investment.

In this view, interest rates are used to target inflation expectation, using unemployment as a tool wrt a rule that is based on a presumption of a natural rate of unemployment defined as "full employment." This creates a buffer stock of unemployed, which implies permanent idle resources. Idle resources are inefficient and wasteful, which economists agree should be avoided if possible.

This is admittedly somewhat of an oversimplification since the monetarist position has evolved since Friedman developed it, but it gives the basic idea as a heuristic device.

Keynesians dispute QTM. For a Post Keynesian explanation of QTM, see John T. Harvey, Money Growth Does Not Cause Inflation (Forbes, May 14, 2011).

Fiscalism and MMT

For fiscalists, employment is of primary concern. Y (income) is the independent variable in PY = MV, changes in which affect effective demand. So fiscalists hold that Y needs to be controlled through fiscal policy, which affects effective demand. Effective demand draws forth investment to meet profit opportunity, and effective demand is income-dependent, since consumption cannot be funded by drawing down savings, selling assets, or financed by borrowing sustainably. If supply and demand are stabilized at optimal resource use, they unemployment is reduced.

The holy grail of macroeconomics is full employment along with price stability, which implies highly efficient use of resources while controlling price level.

In the first place, MMT rejects the monetarist explanation virtually in toto, claiming that it is based on an incorrect view of actual operations of the Treasury, central bank, and commercial banking, and how they interact. Secondly, MMT explains how to succeed in the quest for the holy grail through employment of the sectoral balance approach developed by Wynne Godley and functional finance developed by Abba Lerner. The thrust of this approach is to maintain effective demand sufficient for purchase of production (supply) at full employment by offsetting non-government saving desire with the currency issuer's fiscal balance. This stabilizes aggregate demand and aggregate supply at full employment (adjusting aggregate demand wrt changes in population and productivity) without risking inflation arising owing to excessive demand.

Note that this does not apply to price level rising due to supply shock, such as an oil crisis provoked by a cartel exerting a monopoly, or shortage of real resources., e.g. due to natural disaster, war, or climate. This is a separate issue and must be addressed differently according to MMT.

In a non-convertible floating rate monetary system, the currency issuer is not constrained operationally. The only constraint is real resources. If effective demand outruns the capacity of the economy to expand to meet it, then inflation will result. If effective demand falls short of the capacity of the economy to produce at full employment, then the economy will contract, an output gap open, and unemployment will rise.

This view is based on a Treasury-based monetary regime, in which money is created through currency issuance mediated by government fiscal expenditure. Issuance of Treasury securities to offset deficits functions as a reserve drain, which functions as a monetary operation that enables the central bank to hit its target rate rather than being a fiscal operation involving financing. Similarly, taxes are seen not as a funding operation for government expenditure, but as a means to withdraw non-government net financial assets created government expenditure, in order to control effective demand and thereby reduce inflationary pressure as needed iaw the sectoral balance approach and functional finance.

This view is quite the opposite of the credit-based monetary presumptions of monetarists, which MMT regards as appropriate to a convertible fixed rate regime like the gold standard but not to the current non-convertible floating rate system that began when President Nixon shut the gold window on August 15, 1971, and was later adopted by most nations, excepting those that pegged their currencies, ran currency boards, or gave up currency sovereignty as did members of the European Monetary Union in adopting the euro as a common currency.

It is important to note that MMT economists are NOT recommending the adoption of a Treasury-based monetary system. Rather, they are asserting that the present monetary system is already Treasury-based operationally, even when governments choose to impose political restraints that mimic obsolete practices and create the impression that these are operationally necessary.

MMT also recommends an employer assurance program (ELR, JG) to create a buffer stock of employed that the private sector can draw on as needed. This reduces idle resources and presents the possibility of achieving actual full employment (allowing 2% for transitional) along with price stability, which monetarism presumes inflationary. The ELR program also establishes a wage floor as price anchor for price stability.

(This post grew out of a short comment I made at Warren Mosler's blog in response to a question asked by Mario. Hat tip to Mario for bringing it up.)


PaulJ said...


beowulf said...

"MV = PY is an identity that means the amount of money (M) multiplied by the turnover of money (V for velocity) is identical to the amount of economic activity (Y) multiplied by the price level (P)... taxes are seen not as a funding operation for government expenditure, but as a means to withdraw non-government net financial assets..."

The Fed can use its fee schedule on all FRS transactions as an adjustable "turnover tax" (with net earnings going to Tsy) that drains M and V.

Mario said...

wow. incredible. I'm passing this one around.

The hat tip goes to Neil as well for initially bringing up the equation too.

GREAT stuff here.

Honestly Tom, after a few typo-checks, and maybe a few scholarly references and discussions, this could VERY EASILY be submitted to economic panels and research institutions to further the MMT platform. Quite seriously I mean that. I encourage you to consider polishing this even further and submitting this "around town." This is huge and great stuff.

Tom Hickey said...

Thanks, beowulf.

I would prefer to see the Fed set the overnight rate to zero and use this on a sliding scale wrt to the sectoral balance approach and functional finance than to have them continue attempting to micromanage the economy using interest rates and a Taylor rule.

It's also more practical than using tax policy, especially given the sorry state of the highly partisan political process in which parties jockey for power instead of governing.

Adam said...

I don't think the credit based approached is wrong. Just incomplete.

I think MMT describes both horizontal (credit-money) and vertical (net financial assets) money.

Credit money however is not the way that monetarists describes it. Credit money is endogenous to the banking sector based on the desires of the customers and the potential profitability of the banks (loan officer analysis). Basically what Minsky has hypothesized. Minsky actually went further stating that it is inherently unstable.

For credit-based approaches it is not Monetarism vs MMT. But rather Monetarism vs Monetary Circuit Theory (Steve Keen and Richard Werner's approach)

Detroit Dan said...

Good post. Thanks.

You might add the concepts of "spending multiplier" and "money multipler", which seem to be integral to the monetarist and fiscalist theories. You covered this to some extent, but it might help to spell out the contesting multipliers...

Tom Hickey said...

Thanks, Dan. Of course, there is much more to say about this. My post was inspired by Neil Wilson's comment at Warren's on reversing the MV = PY identity to PY = MV, thereby switching the independent and dependent variables. I riffed off that.

Dan Kervick said...

Very clear. Well done, Tom.

Oliver said...

Good summary!

One quibble:

You say: For fiscalists, employment is of primary concern.

It seems you're conflating the normative and the descriptive aspects of MMT with this sentence. Could it not be said that fiscalsim is a theory about the way our current monetary system operates. And that understanding this theory, allows those who are concerned about employment to bypass the supposed constraints inherent in Monetarist and Classical theories.

In theory, one could believe in the Chartalist analysis but not give a damn about unemployment. But, if Chartalism were the accepted paradigm, such an attitude could be better exposed for what it is, namely a complete disregard for human welfare as those who hold it could no longer hide behind the veil of the goddess NAIRU.

circuit said...

Good post. I especially like the last part and the paragraph where you highlight that the system is *already* Treasury-based from an ops standpoint. I've noticed some confusion surrounding this issue in the comments following one of my posts. Perhaps Olivier is right that confusion exists between the descriptive and prescriptive aspects of the approach.

Mario said...

Agreed completely Oliver. Great way of putting it.

Tom Hickey said...

Oliver, the basic challenge of macro policy-wise is reconciling employment and inflation.

in this broad brush summary, I was thinking of it terms of Friedman v. Keynes. So the monetarists are the Friedmanites and the fiscalists are the Keynesians. There are different varieties of both, with monetarists chiefly concerned with stability of purchasing power of the currency and fiscalists primarily concerned with stability of employment. I am representing MMT as a school of fiscalism, one among several others.

MMT reconciles both positions in its claim to be able to achieve full employment defined rather strictly with price stability. Previously most macro folks saw as impossible without redefining key terms to fit the theory.

circuit said...

As a follow-up to my previous comment, I should specify that the work of Scott Fullwiler (eg. primer on CB ops and academic papers) and Randy Wray (eg. papers and the NEP MMT primer) are good resources for understanding the descriptive aspects. However, it's the policy proposals associated with the approach that seem somewhat scattered (in Levy papers, blog posts, Warren Mosler's campaign proposals, etc). Myself, I always equated the core MMT proposals more or less with ELR and the Kansas City Rule (ie. zero nominal interest rate that Tom mentions above).

@ Tom: Do you know of a good link/document that you believe adequately outlines the other (more specific) set of policy prescriptions associated with MMT or advanced by modern money economists? (Though, perhaps you're like me and think the descriptive is/should be the real focus, thus allowing for discussions on the different possible policy options rather than on specific proposals)

Mario said...

how does MMT respond to the fact that a zero rate would hurt savers and help speculators? B/c I know that many people feel that way about 0 rates these days and QE, etc. and frankly I don't have a good response.


Chewitup said...

Once everyone learns the operational aspects of MMT, the prescriptions can be debated from all points of view as one defines public purpose. Randy has stated in the NEP blog that MMT can be used by progressives, libertarians, Austrians, conservatives, etc. The important part is getting everyone on the same page regarding the realities of our monetary system.

Ralph Musgrave said...

Bo**ocks to fiscal and bo**ocks to monetary. Any monetary policy on its own is distortionary and is thus no good or at least sub-optimal. For example interest rate adjustments (the classic monetary policy) work only via entities that are significantly reliant on loan rather than equity finance (that’s assuming that interest rate adjustments work at all, which is debatable). What’s the logic in boosting an economy via one set of households and firms, but not the rest?

As to fiscal, no one can agree on the extent of crowding out, so that’s not much use. Moreover, what’s the point in government borrowing stuff (i.e. money) which it can produce itself at no cost? Madness.

The best policy is the simple Abba Lerner / MMT combination of fiscal and monetary, i.e. print money and spend it in a recession (and/or cut taxes). And when inflation looms, do the opposite.

And perhaps that answers Mario’s question just above about interest rates. I.e. the answer is to ignore interest rates: let the market determine interest rates. Governments should not tamper with them.

Mario said...

interesting Ralph.

so even at zero rates banks could set their own competitive rates for savers? I just assumed they would let them sit at some algorithmic multiple from zero that only moves when the base rate moves. I guess the other possibility is that the support rate would be moving instead from then on, so savings rates theoretically would also move in relation to the support rate just as they used to off the ffr.

Tom Hickey said...

circuit, the only comprehensive account of MMT that I know of is Randy's book Understanding Modern Money.

MMT is 1) a operational description of the modern money system, 2) a macro theory, 3. policy options based on 1 and 2.

BTW, the ELR is part of MMT's macro theory for achieving full employment and price stability (buffer of employed/price anchor), not a policy prescription based on it.

Tom Hickey said...

Mario: "how does MMT respond to the fact that a zero rate would hurt savers and help speculators? B/c I know that many people feel that way about 0 rates these days and QE, etc. and frankly I don't have a good response."

Warren's policy proposal is setting the overnight rate to zero and letting excess reserves accumulate and eliminating issuance of anything longer than a 3 mo T-bill.

That way the market would set rates instead of a small group of unelected and unaccountable technocrats, which is anti-democratic and anti-capitalistic.

Mario said...


okay. So does that mean that rates would ever fluctuate at all? And if they did, would it be b/c of "the market" for 3 month bills supply/demand?

Would support rate fluctuations factor at all into savings rates in that scenario?

Dan said...

Would you clarify why GDP = Y rather than PY? My understanding has been that P is the average price per transaction and Y is the number of transactions. Then Y would relate more closely to demand which is of primary interest to MMT.

circuit said...

Wray '98. Classic. Tks.

Tom Hickey said...

Mario, interest rates fluctuate with demand. The primary demand comes from demand for capital. The prevailing rate at which firms can borrow is the cost of capital, and businesses use cost of capital as factor in decisions to invest.

The problem with the cb setting interest rates to control inflation is that this operates through the cost of capital instead of letting the market determine it. That is inefficient and leads to distortions.

MMT proposes using fiscal policy to control both employment and price stability, and Warren's proposal is to let the market set the cost of capital based on demand.

Tom Hickey said...

Dan: Would you clarify why GDP = Y rather than PY? My understanding has been that P is the average price per transaction and Y is the number of transactions. Then Y would relate more closely to demand which is of primary interest to MMT.

MV = PY = GDP (nominal). Nominal GDP is sometimes specified as NGDP.

This was initially written MV = PT instead, where expenditure on all transactions is summed. This would required multiplying all transactions by the price. Obviously, this is impossible to comput for an entire economy. So an average was used instead. But even so, there is no reported data on the volume of transactions across sectors. So PT was abandoned.

IT is simpler and more practical to aggregate expenditure across sectors and adjust for inflation to get real GDP (RGDP). So PY was substituted for PT. PY in MV = PY is price level times real output, where price level is usually CPI. Some write MV = Py to show that y = RGDP.

DanD2 said...

Tom Hickey:
Wrt Dan's question, I think it arises from the two different uses of the symbol, Y, in the first three paragraphs of your post. In the first two paragraphs, Y is the number of transactions occurring in a given time period (widgits/time). In the third, Y is nomGDP ($/time).

Tom Hickey said...

Dan, the two equations are substantially the same, but only differ in computation. There are different methods of computation of economic activity, e.g., transactions, income, expenditure.

MV = PT, called the equation of exchange, was developed by Irving Fisher in 1911, using transactions. Later, the equation was simplified to MV = PY, using the chiefly the expenditure method. See Wikipedia — The Quantity Theory for a summary of the history.

Friedman used MV = PY in his development of the monetarist theory. Subsequent monetarists worked within this framework.

Keynes rejected the quantity theory because it presumes that money is neutral (changes in the money supply have only nominal effects). Keynes argued that changes in money supply work through effects on aggregate demand, which impacts the real economy. E. g., if money increase is saved, it doesn't affect NAD. Subsequent "fiscalists" worked in terms of aggregate demand effects.