QE is not Inflationary – Just Ineffective

In the US, the Fed has just commenced another round of quantitative easing. This is despite the absence of any evidence to suggest that previous implementations of the policy – in Japan, the UK or the US – have had any significant impact on demand, output and employment. At the same time, some critics of quantitative easing claim that the policy will be inflationary. But this claim also defies both theoretical considerations and the empirical evidence. The policy is not inflationary, just impotent.

The Fed’s decision to persist with the policy probably says more about the neo-liberal distaste for discretionary fiscal policy than any confidence that QE2 will be more successful than its predecessor.

An earlier discussion of quantitative easing can be found here. The policy involves the Fed purchasing longer term bonds and paying for them by crediting the reserve accounts of the commercial banks.

Since quantitative easing simply swaps one type of asset for another, it does not alter the net financial assets of the non-government sector in any significant way. It merely alters their composition. There is no significant boost to the purchasing power of the non-government sector as a whole. The sector has less bonds but more cash (bank reserves). Strictly speaking, the sector has relinquished higher interest earning bonds for lower interest earning reserves (reducing purchasing power), but sold them at a price somewhat higher as a result of the Fed’s action (increasing purchasing power). Overall, there may be some minor purchasing power effect, but only a very small one, and the direction of the effect is unclear.

There is little reason to believe that this alteration in the composition of the non-government sector’s net financial assets will encourage greater spending in the real economy. Prior to the quantitative easing, individual holders of the bonds were perfectly at liberty to exchange their bonds for cash to make expenditures if they had so desired. They didn’t want to spend on real goods and services, so had no reason to convert to cash. Under quantitative easing, they became willing to convert to cash, but it was not because they wanted to spend but because the Fed had pushed up the price of longer term bonds, making some non-government sector participants willing to alter the composition of their savings, but not necessarily the level of their savings.

For quantitative easing to have a positive impact on aggregate demand, and hence potentially cause inflation if and when the economy’s productive capacity is reached, the policy would need to induce a change in the saving behavior of the non-government sector. Assuming this is the intention of the policy, the Fed might conceivably be relying on one or other of the following mechanisms:

(i) The expansion of reserves inducing greater lending and a multiplied increase in bank deposits (this is actually a false hope and not one that modern central bankers appear to believe in but worth dispelling because the flawed idea remains ubiquitous in textbooks);

(ii) A reduction of longer term interest rates stimulating aggregate demand through a positive impact on private productive investment;

(iii) Perceived wealth effects stimulating aggregate demand.

Money Multiplier

Mechanism (i) is the fictitious ‘money multiplier’ effect, which rests on the false premise that banks need sufficient reserves prior to lending. By this reasoning, the expansion of bank reserves associated with quantitative easing supposedly gives commercial banks more capacity to extend loans.

But this is not how banking in the current system works. Banks do not require prior reserves in order to make loans. They require credit-worthy borrowers (and sufficient capital). They need to see that the prospective borrower has a realistic prospect of repaying the loan. If the bank considers the borrower a good risk, it will extend the loan by crediting a bank account. In this way, the loan creates a new deposit: loans create deposits, not the other way round. If at the end of the day the bank is short of reserves, it simply borrows them from another bank with excess reserves, or, if the banking system as a whole is in deficit (for example, because tax payments exceeded government expenditure over the period), the bank borrows the necessary reserves from the Fed. For a simple explanation of this point, see this post.

In assessing loan applications, the bank (assuming sufficient capitalization) will focus on the credit-worthiness of the loan applicant. If the prospective borrower is a firm, the firm’s intended productive investment will need to appear sufficiently profitable to earn for the firm more than the interest owed on the loan. If the prospective borrower is a household (most commonly someone wanting to take out a mortgage on a home), the household will need to demonstrate to the satisfaction of the bank its capacity to earn sufficient income, primarily through employment, to repay the loan.

What matters in assessing the loan application is the credit-worthiness of the prospective borrower in the estimation of the bank. The bank’s assessment is influenced by the general economic situation. Lending requirements have tightened because general economic prospects are poor. Banks can’t be sure of household members retaining their jobs or finding new jobs and can’t be confident of future demand conditions in firms’ product markets.

Now, in aggregate, quantitative easing does nothing to alter the credit-worthiness of borrowers. As was mentioned at the outset, it causes no (or very little) alteration in the net financial assets of the non-government sector. The non-government sector has relinquished some bonds (a decrease in net financial assets) in exchange for bank reserves (an increase in net financial assets). The purchasing power and capacity to service debt of the non-government sector has, in aggregate, remained more or less unchanged.

Further, the sellers of the bonds were, by definition, bondholders, and the bonds came from that part of their income designated for saving, not spending. So they have switched their form of saving from bonds to cash to the extent that they participated in the Fed’s operation. Unless these people now decide that they want to save less and spend more, there will be no effect on the broader money supply. All that happens is that the reserves of the commercial banks mount. There is no impetus for the extra reserves to be lent out for productive purposes. They will just sit in the banks’ accounts with the Fed. And this is what has been happening.

As long as the net financial assets of the non-government sector remain unchanged, the only hope for a stimulus to aggregate demand is through an alteration in the saving (and spending) plans of the non-government sector. If this sector suddenly decided it wanted to spend more and save less, there would be more demand. But there is a reason the non-government sector is saving so much. The sector over-extended itself in the lead up to the crisis, becoming highly indebted under the false perception that real estate and Wall Street bubbles represented a real increase in wealth rather than fictitious value. Once these bubbles burst, people lost a lot of nominal wealth but were still stuck with their debt. This is why the non-government sector needs to net save. As a matter of accounting, the government must run deficits if this net saving is to occur.

So there is no reason to expect an expansion of bank reserves to induce an increase in private lending and a multiplied expansion of the broader money supply. But, just for the sake of argument, it may be instructive to suppose – contrary to operational reality, logic, and empirical evidence – that the expansion of reserves did somehow result in more loans and a multiplied expansion of the broader money supply, and consider whether this would be inflationary. The answer is that it would not be inflationary unless and until the debt-financed expenditures tested the supply-limits of the economy. Any demand – whether government, private or external – can cause inflation if the economy is pressing up against its supply limits. But when there is high unemployment and excess capacity, competition compels firms to respond to increased demand through expansion of output and employment at current prices. Otherwise the firm will miss out on potential sales and lose market share. If we are frightened of the inflationary effect of an expansion of loans to fund domestic private expenditure, then by the same logic we should be frightened of an expansion of export demand. If the level of economic activity was such that an increase in demand will be inflationary, then an increase in any form of demand will be inflationary. But this is not at all a danger when there is mass unemployment and excess capacity. And if and when it became a danger, due to a strong and sustained rise in demand that tested the supply limits of the economy, the Fed – since we have assumed in this paragraph for sake of argument that this tremendous boom has grown out of the Fed’s strategy of quantitative easing – would merely need to reverse its operation by selling back longer term bonds in exchange for bank reserves. By the same logic, this contraction in reserves would induce a multiplied contraction in private lending and broader money.

But that last paragraph was just for the sake of argument and its premise is actually a complete fiction. The expansion of reserves achieved through quantitative easing has no such positive impact on private lending and aggregate demand, and therefore poses no inflation risk, even potentially as an indirect boost to aggregate demand.

Private Investment and Interest Rates

Another hope of the Fed may be that reducing longer term interest rates will stimulate private investment and aggregate demand. Such an effect relies on a highly unreliable transmission mechanism. First, other components of aggregate demand may be dampened even if private investment is stimulated, leaving the overall effects on aggregate demand uncertain. Most notably, lower long-term interest rates reduce the interest income of savers who will reduce consumption in accordance with their saving plans. Second, the supposed inverse relationship between private investment and longer term interest rates ignores the revenue side. Although lower interest rates reduce the cost of borrowing, weak demand conditions in product markets depress revenue and profit expectations. Third, the alleged inverse relationship between private investment and longer term interest rates has been shown not to hold on logical grounds. This is a chief lesson of the capital debates.

There are some signs that mainstream policymakers increasingly understand the unreliability or invalidity of effects (i) and (ii). With this in mind, some people think the Fed is pinning its hopes on a third effect.

Wealth Effect

The Fed may be hoping that another Wall Street bubble will create a (fictitious) wealth effect that might stimulate demand. That is, if the non-government sector, in aggregate, feels wealthier in response to another Wall Street bubble, it may be lulled into taking on more debt in order to spend more and kickstart the real economy. There is little evidence that the quantitative easing to date has produced a wealth effect. Even if it could, it would be a foolish use of policy, because any boost to demand achieved in this way would be unsustainable due to the heavy indebtedness of the non-government.


The bottom line is that the private sector cannot sustainably lead the economy out of a balance-sheet recession. If policymakers want the private sector to drive recovery, they need to arrange for an orderly cancellation and/or significant write down of private debts. This would reduce the need for non-government net saving and free up income for private expenditures.

If, on the other hand, policymakers want to prevent creditors from taking losses, as appears to be the case, then there is only one viable solution: strong and sustained fiscal stimulus targeted directly at the real economy.

Monetary policy is ineffective in a balance-sheet recession. Banks will not extend loans to poor credit risks.

Fiscal policy targeted at job creation is more effective. Large and sustained budget deficits aimed at employment creation result in higher income and enable non-government net saving to repair balance sheets alongside stronger demand conditions. Once balance sheets are in better shape, the non-government sector will be in a position to drive economic activity, and banks will be more willing to lend.


6 thoughts on “QE is not Inflationary – Just Ineffective

  1. So, just to make sure I understood what’s going on. The QE is a horizontal transaction in which no new NFAs are created. So, the reserves that Fed is using to buy the bonds are not in G in the G-T=S-I+M-X; they do not involve any new spending by the government (and they don’t need to be authorized by Congress either). However, they are still created ex nihilo by the government sector (the CB). So, if my understanding is correct, horizontal transactions include credit money creation by the banks, swaps of financial assets between agents in the private sector and swaps of financial assets done by the government itself through the CB?

  2. Peter D, the cb creates reserves by typing numbers into spreadsheets, thereby marking up accounts. It reserves the operation by using minus signs. There is nothing backing the reserves. That is what ex nihilo means.

    However, actual transfers having financial meaning do occur, and they are revealed in the accounting. One can think of it as the cb issuing reserves and loaning them to Treasury for settlement of its checks. Treasury then issues tsys and gives them to the cb to settle the loan. The cb sells the tsys for reserves. This way the cb recoups the reserves it loaned to Treasury via the sale of tsys. Think of it as the reserves that came out of the hopper going back into it.

    Notice that all accounts balance, but in the process reserves were created by the cb and tsys by Treasury ex nihilo in order to make the accounts balance. This is a lot simpler to understand if one thinks of it in terms of balancing accounts.

    Since this is fiscal and changes nongovernment net financial assets, it is under the direct control of Congress through appropriations and taxation. So it is not just the cb and Treasury creating money as they please.

    If the cb wants to buy tsys through QE, it just uses reserves to do it. You can conceptualize this as the “same” reserves being used that it got back from the sale of tsys if you want, although that is meaningless. Similarly, if the cb sells the tsys purchased through QE back into the market, then it gets the reserves used to buy them back again. It is just a matter of balancing accounts wrt to transfers. There is no change in nongovernment NFA, so it’s monetary, not fiscal, and doesn’t affect (G-T).

    Essentially the same thing that happens when banks create deposits “ex nihilo” through loans. Banks don’t loans out deposits. Nor do they loan out reserves. They need reserves for settlement, but that comes afterward. All this is just moving numbers around on spreadsheets, which shows up on bank balances, checking and credit card account statements, etc. Relatively little cash is used to settle transactions now.

    Final settlement of all transactions occurs in either bank reserves or cash, unless the exchange is intrabank, and bank reserves are simply a settlement vehicle used in balancing accounts. This is handled through the interbank system, in the US the FRS. It’s just shifting numbers in accounts among banks in settlement of transactions occurring in the economy. Bank reserves shift around and no change in nongovernment NFA takes place.

    In summary: On one hand, loans remain on the banks’ books as accounts receivable until they are paid off, which also extinguishes the deposit created by the loan as an account payable. On the other hand, reserves remain in the system either as reserves or saved as tsys until they are withdraw by taxation. So nongovernment net financial assets (savings) created by deficits remain until withdrawn by corresponding surpluses. In the US, deficits are offset $-4-$ by tsy issuance, thus the national debt is the saving of net financial assets by nongovernment.

    “Vertical” generally applies to changes in nongovernment net financial assets through fiscal operations. “Horizontal” applies to banking operations that do not affect NFA, i.e., net to zero.

    Note, however, that the cb supposedly only engages in monetary ops, which do not affect fiscal. But when the cb purchases tsys it does transfer the interest that would have been paid to non-government. Since interest on tsys is fiscal in that it affects nongovernment net financial assets, QE does have fiscal implications by withdrawing interest that would have been paid to nongovernment. Similarly, should the Fed sell its tsy holdings to unwind QE, it will be returning that fiscal stimulus to nongovernment.

  3. Tom, I think you confirmed my understanding, unless I am missing something, that QE is an example of a horizontal transaction (with a little vertical effect having to do with interest) done between the government sector (represented by the CB) and the private sector.
    So, to generalize a bit further, the distinction between vertical and horizontal transactions is not about government-to-private vs. private-to-private but rather fiscal vs. asset-swap transactions, where “asset-swap” is understood broadly (for example, a bank loan is swapping an asset of deposit for an asset of loan)

  4. Peter D., vertical and horizontal are chiefly about money creation. Generally speaking, MMT’ers avoid the use of “money” as ambiguous, but Randy Wray has a short paper entitled Money that delineates the concept and is well worth reading.

    “Vertical money” is “government money” or “government credit.” Government fiscal operations create (inject) and uncreate (withdraw) net financial assets wrt nongovernment. Government credit is essentially a tax credit, but it also serves to meet other financial obligations to government such as fees and fines. The nongovernment liability corresponding to government credit is what nongovernment owes to government, that is, taxes, fees, fines, etc. Government can create more or less credits than it creates nongovernment liabilities to itself, thereby effecting a net balance that is unequal to zero. Because taxes far outweigh fees and fines, government is in control of the net it wishes to set through its budget.

    Horizontal money” is “bank money” or “bank credit” (loans equal deposits), so it nets to zero. Because banks are public-private partnerships banks can create credit that is denominated in the currency of issue. They can do this because they have assess to the interbank settlement system that operates on the basis of reserves that are created by government through the agency of the central bank.

    The connection between vertical and the horizontal is high powered money (reserves and physical currency), which is under the control of government and which is necessary for final settlement of all monetary transactions.

    In modern economies, governments are generally given monopoly powers wrt to the currency of issue by the national constitution. Governments then delegate some of this power to its agencies, such as Treasury and the cb, and also to banks as public/private partnerships. Banks are chartered, for example, and are required to abide strictly by regulations. As a result, governments have complete control of the national currency, to the degree they wish to exercise it.

    As you note, fiscal ops alter nongovernment NFA and swaps don’t. Swaps don’t since they net to zero. The significant matter is that vertical ops are fiscal and affect the net position of nongovernment, while horizontal ops are monetary or financial and don’t affect the net. Of course, horizontal operations can and do affect real assets.

    All this is important to MMT because of the sectoral balances. MMT holds that fiscal operations affecting nongovernment net financial assets influence demand, e.g., budgetary deficits offset demand leakage to increased saving desire or CAD since deficits increase nongovernment NFA. What this means is that the increase in nongovernment financial assets (saving) is not offset by a corresponding nongovernment liability (debt). Similarly, a budget surplus withdraws nongovernment NFA and reduces demand unless nongovernment prefers to either draw down or take on debt.

    MMT recommends using vertical injection and withdrawal to influence demand fiscally by altering nongovernment NFA. This is the basis of functional finance as a policy instrument for influencing nominal aggregate demand directly in a targeted way through expenditure and taxation.

    On the other hand, monetarism involves attempting to influence the balance of saving and investment, using the interest rate (overnight price of reserves) to shift saving/investment preferences. This is a blunt instrument in any case since it can’t be targeted, and it is in op at the zero bound. Formerly, monetarism attempted to use the purported money multiplier to influence the money supply through bank credit by changing the quantity of reserves, but it was found to be ineffective and largely abandoned.

  5. Great explanations, Tom. Thanks for the link to the Randall Wray paper as well. I agree that it is very much worth reading.

  6. Hey Heteconomist,
    Thanks, on a related note, For the Fed it is inflate or die. If the Fed dies, the economy suffers a period of adjustment as too big to fail financial predators take their losses and capitalism reorganizes with financial institutions that actually serve the nation and the markets. If the Fed continues to have its way and inflates the currency, everyone loses, as all savers and investors and real capitalists will see their wealth destroyed.

    The Fed is a modern day Babel. Babel and Babylon signify total confusion. Usury banking originated in Babylon, which symbolizes the futile attempt of the oligarchs to enslave mankind. Under the Fed the dollar has already lost 97 percent of its purchasing power. Under QE to infinity it will be destroyed.

    This Babel of financial insanity is destined to fall. Hidden in plain sight in the Bible is a prophecy that at an appointed time in the future, there will be a great awakening and world wide debtors revolt.

    The Vision of Habakkuk, written around 620 B.C., reveals that this great awakening of the debtors is the key event to the collapse of a secretive world-controlling system that has been plundering the wealth of nations throughout history.

    Habakkuk was instructed to write the vision in big letters on tablets, making it plain (easy to understand) so that all that read it can run with it, communicating the message to others:

    Write the Vision and make it plain
    upon tablets, that he may run that reads it.
    For the Vision is for an appointed time.
    And it is not a lie. Wait for it, because it
    Will surely come to pass, and not delay
    When the time is right.

    Here is the core of the prophecy announcing the great awakening:

    Woe to the proud, who has taken what is not his,
    Making himself rich with loans.
    Suddenly the debtors will awake, as a man out of sleep,
    And they will bite him back, in an instant.
    As he has ruined the nations, so the nations will ruin him.

    The prophecy further declares that when you first hear of it, that you will not believe it, but that it is not a lie. When this event does occur, it will be a wonderful providential blessing for all mankind.

    It is prophesied and it will happen.
    Keep up the good work

Comments are closed.