Japanese economist Richard Koo recently (July 9, 2013) published his latest report on the world economy – Japan, US, and Europe face different issues – which updates some of the latest data available from the economies listed in the title. I am sorry that I cannot link to the Report as it is a subscription service (thanks to Antoine for my copy). I discussed some of Richard Koo’s ideas and how they sat with Modern Monetary Theory (MMT) concepts in this 2009 blog – Balance sheet recessions and democracy. While the basic concept of a balance sheet recession is important to grasp and the policy prescriptions that flow from it clearly point to the need for more fiscal stimulus, once you dig a little deeper into Koo’s conceptual framework you realise that he is very mainstream – more insightful than the average mainstream economist, who typically fails to even grasp the reality of the current situation, but mainstream nonetheless. And that means there are some things in his theoretical framework that are plain wrong when applied to a modern monetary economy
The background blogs, which provide the essentials to understand the argument summarised today are as follows:
- Money multiplier and other myths
- Money multiplier – missing feared dead
- Building bank reserves will not expand credit
- Building bank reserves is not inflationary
- Lending is capital- not reserve-constrained
- The role of bank deposits in Modern Monetary Theory
In my discussion of Richard Koo’s (RK) Report I will not highlight areas of agreement but just focus on the areas of disagreement. That is a time-saving strategy but you should not conclude from the resulting negativity that there is nothing in the Report that I agree with.
RK discusses the problems that might arise if “private loan demand” recovers too quickly and concludes that “Central banks would prefer gradual recovery” because a “(g)radual pick-up in loan demand would allow Fed to mop up excess reserves over time”.
The logic presented is as follows:
Under ordinary circumstances, an economic recovery cannot come too soon. But authorities that have engaged in QE … An abrupt rebound in demand for funds could prompt US banks—which hold excess reserves equal to 16 times statutory reserves—to increase lending suddenly. The Fed would then have to mop up excess reserves quickly out of concern for inflation and asset price bubbles.
Doing so at a time when the private sector was seeking to borrow could cause interest rates to rise sharply, with negative implications for the economic recovery.
The situation is made worse since the assets the Fed would have to sell are mainly longer-term bonds. Unloading those on the market could push long-term interest rates dramatically higher and harm the housing sector, the engine for the ongoing recovery.
A more gradual pickup, according to RK will allow the central bank to more slowly absorb excess reserves with less impact on interest rates.
This discussion implies that there is a huge stockpile of money that the banks have available (the reserves) which they might just lend out quickly. That impression is false.
First, banks do not need reserves to lend? The mainstream view is that reserves are deposits that haven’t yet been loaned.
But the reality is that banks do not lend reserves generally – that is, to customers. They might loan them to other banks in the system, which have accounts with the relevant central bank and who are short of the funds required to cover their obligations to the clearing system. But they certainly do not loan out funds from these reserve accounts at the central bank to you and me.
This is an area of considerable misunderstanding. What actually is the role of bank reserves?
Commercial banks are required to keep reserve accounts at the central bank. These reserves are liabilities of the central bank and function to ensure the payments (or settlements) system functions smoothly. That system relates to the millions of transactions that occur daily between banks as cheques are tendered by citizens and firms and more.
Without a coherent system of reserves, banks could easily find themselves unable to fund another bank’s demands relating to cheques drawn on customer accounts, for example.
Depending on the institutional arrangements (which relate to timing), all central banks stand by to provide any reserves that are required by the system to ensure that all the payments settle. The central bank charges a rate on their lending in this case which may penalise banks that continually draw on the so-called “discount window”.
Commercial banks thus will have a reserve management area within their organisations to monitor on a daily basis their status and to seek ways to minimise the costs of maintaining the reserves that are necessary to ensure a smooth payments system.
The interbank market (say the federal funds market in the US) functions to shuffle the reserve balances that the member (private) banks keep with the central bank to ensure that each of these banks can meet their reserve targets which might be simply zero balances at the end of the “day”.
I have put “day” in inverted commas because we should think that the central bank polices its reserve requirements per day. They requirements are usually expressed over some period of weeks rather than days and are averages. But that is a complication we can avoid here.
So banks can trade the balances in these reserve accounts between themselves on a commercial basis but in doing so cannot increase or reduce the volume of reserves in the system. Only government to non-government transactions (which in MMT are termed vertical transactions) can change the net reserve position.
The point is that all transactions between non-government entities net to zero (and so cannot alter the volume of overall reserves). I explain that in more detail including the implications of that point in the trilogy of blogs – Deficit spending 101 – Part 1 – Deficit spending 101 – Part 2 – Deficit spending 101 – Part 3.
It is in this context that RK talks about mopping up the reserves in the system. What he is referring to is that under quantitative easing, the central bank swapped financial assets for reserves and if it wants to eliminate the reserves that were created it can reverse that transaction.
In the same way that the bond purchases that were made on a significant scale reduced interest rates in the maturity segment of the yield curve that the bonds existed (for example, 10-year bonds), it is argued that the sale of them again will depress bond prices and push up yields accordingly.
Fixed-interest assets such as government bonds can trade freely on secondary bond markets after being issued by the authorities in the primary market but their yields then vary inversely with their prices, which, in turn, depends on the relative demand for them.
His fear is that the banks might hurry the central bank into conducting this reversal if they lend these reserves out too quickly.
The flaw in his argument is that the banks do not lend the reserves out. It doesn’t matter how large the balances are there is no increased capacity to lend on behalf of the banks.
Once we realise that point then we can separate the two strands of his argument. An “abrupt rebound in demand for funds” would only occur if there was an abrupt increase in the number of credit-worthy customers willing to borrow.
If that was to manifest, it wouldn’t matter if there were huge reserve balances or zero reserve balances. The banks would start to lend rapidly and the reserves would be found to cover the requirements of the payments system defined above.
The important point is that when a bank originates a loan to a firm or a household it is not lending reserves. Bank lending is not easier if there are more reserves just as it is not harder if there are less. Bank reserves do not fund money creation in the way that mainstream monetary economics theory, which places the money multiplier and fractional-reserve deposits at the centre of the analysis, has it.
Modern Monetary Theory (MMT) notes that bank loans create deposits not the other way around. Reserve balances have nothing to do with this – they are part of the banking system that ensure financial stability.
These loans are made independent of their reserve positions. So while the bank organisation will include a reserve management division it also will have a loan division. The two are functionally separate and the latter will not correspond with the former prior to making loans to appropriate credit-worthy customers.
Depending on the way the central bank accounts for commercial bank reserves, the banks will seek funds to ensure they have the required reserves in the relevant accounting period.
They can borrow from each other in the interbank market but if the system overall is short of reserves these “horizontal” transactions will not add the required reserves. In these cases, the bank will sell bonds back to the central bank or borrow outright through the device called the “discount window”. There is typically a penalty for using this source of funds.
At the individual bank level, certainly the “price of reserves” will play some role in the credit department’s decision to loan funds. But the reserve position per se will not matter. So as long as the margin between the return on the loan and the rate they would have to borrow from the central bank through the discount window is sufficient, the bank will lend.
At present, there are large reserve balances, which alter the price of reserves. But the idea that reserve balances are required initially to “finance” bank balance sheet expansion via rising excess reserves is inapplicable.
To summarise, a bank’s ability to expand its balance sheet is not constrained by the quantity of reserves it holds or any fractional reserve requirements. The bank expands its balance sheet by lending. Loans create deposits which are then backed by reserves after the fact.
The process of extending loans (credit) which creates new bank liabilities is unrelated to the reserve position of the bank.
The major insight is that any balance sheet expansion which leaves a bank short of the required reserves may affect the return it can expect on the loan as a consequence of the “penalty” rate the central bank might exact through the discount window. But it will never impede the bank’s capacity to effect the loan in the first place.
So it is quite wrong to assume that the central bank can influence the capacity of banks to expand credit by adding or draining reserves to or from the system. Please read the following blogs – Building bank reserves will not expand credit – for further discussion on this point.
RK then concludes that it is very likely that there will only be a “gradual recovery in private loan demand” in Japan, if historical experience is anything to go by.
But then he says that:
When corporate demand for funds finally turned positive in 2006, the BOJ mopped up all the excess reserves, which at the time were equal to about six times statutory reserves.
Had the Bank allowed these reserves to remain in the system, Japan’s money supply could have expanded six-fold, sparking unprecedented inflation. However, such concerns were not realized since the subsequent growth in private loan demand was so gradual.
Once again there is confusion here. Whether the central bank “mopped up all the excess reserves” was moot. The private banks might have still expanded the “money supply” 6-times if loan demand was very strong. The BOJ would have had to wear that or drive up interest rates in an attempt to choke of the demand for loans.
Its liquidity operations (draining reserves) was a quite separate act.
RK considers that US private-sector balance sheets are still being repaired (households and firms are still paying down debt and taking advantage of the near-zero interest rates), which means there is unlikely to be a renewed outbreak of borrowing, of the type that led to the crisis.
Later in the Report he considers the issue of balance sheet recessions and the concept of “fiscal space”. Once again, it is not all plain sailing.
RK considers in relation to the UK and the Eurozone that:
… there is a growing awareness among the authorities that fiscal stimulus is essential during a balance sheet recession. Yet many of them still believe that fiscal stimulus is possible only in countries that have “fiscal space,” and that for countries without it fiscal consolidation is the only option.
“Fiscal space” is meant to describe a country with substantial fiscal leeway that is capable of issuing debt without incident. But many eurozone officials have yet to recognize that this term has very different connotations for countries inside and outside the eurozone.
Which is totally consistent with core MMT.
A balance sheet recession emerges via this sequence of events.
- The private sector builds up massive debt levels to buy property and speculative assets.
- The asset prices rise as demand rises but then eventually the bubble bursts and the private sector is left with declining wealth but huge debt.
- The private sector then start restructuring their balance sheets – and stop borrowing – no matter how low interest rates go.
- All effort is devoted to paying back debt (de-leveraging) and households increase their saving and reduced spending because they become pessimistic about the future.
- A credit crunch emerges – not because there is enough funds but because banks cannot find credit-worthy borrowers to lend to.
- Attempts at pumping liquidity into the banks will fail because they are not reserve-constrained. They are not lending because no-one worthy wants to borrow.
- The faltering spending causes recession and rising unemployment which reinforces the negative outlook.
- With this private contraction (reducing debt, saving) the only way out of the “balance sheet recession” is via public sector deficit spending.
- This deficit support has to be provided for long periods of time while the deleveraging process is occurring.
This is describing a recession that is sourced in the financial markets rather than a more typical real downturn that emerges when firms become negative about future demand and reduce investment. These downturns also require deficit support from government but generally non-government spending rebounds more quickly as there are not legacy issues relating to unsustainable balance sheets.
But digging deeper into RKs argument we find an underlying narrative that is very orthodox.
RK thinks that in the case of a currency-issuing government:
A balance sheet recession requiring fiscal stimulus occurs when there are unborrowed private-sector savings in spite of near zero interest rates. This means the funding for fiscal stimulus to prevent a deflationary spiral already exists in the form of that nation’s unborrowed savings.
Investors faced with the need to invest these unborrowed savings must ultimately buy bonds issued by the government, the sole remaining borrower in a balance sheet recession. That is what provides the necessary “fiscal space” during such a recession.
RK thinks there is some finite pool of private sector savings that can provide the funds for government to spend if no-one else is borrowing them. If these funds are being borrowed by firms, then there is no fiscal space!
However, in reality, rather than in the textbooks, the concept of fiscal space is real, not financial. What does that mean?
Remember the discussion in the blog from last week – The spurious distinction between the short- and long-run.
I noted that the great Polish economist, Michal Kalecki (in his 1943 article Political Aspects of Full Employment Page 348) wrote that:
It may be asked where the public will get the money to lend to the government if they do not curtail their investment and consumption. To understand this process it is best, I think, to imagine for a moment that the government pays its suppliers in government securities. The suppliers will, in general, not retain these securities but put them into circulation while buying other goods and services, and so on, until finally these securities will reach persons or firms which retain them as interest-yielding assets. In any period of time the total increase in government securities in the possession (transitory or final) of persons and firms will be equal to the goods and services sold to the government. Thus what the economy lends to the government are goods and services whose production is ‘financed’ by government securities. In reality the government pays for the services, not in securities, but in cash, but it simultaneously issues securities and so drains the cash off; and this is equivalent to the imaginary process described above.
Kalecki understood this in 1943! Yet, economists still fail to understand it some 70 years later. All the sham institutional frameworks that disguise the essentials, the government just borrows what it has spend in the past.
Kalecki said that a “budget deficit always finances itself”, which is a specific version of the notion that spending brings forth its own saving via income changes.
The loanable funds doctrine considered saving to be finite at any point in time and various users of these funds would compete against each other for access. The interest rate mediated this access and determined who got the funds. In this static environment it is easy to see why they would claim private investment lost out if government attracted the finite saving by offering debt instruments.
However, Keynes formally broke with this view by noting that saving is a function of national income (non-consumption). As income grows so does the pool of saving.
Spending drives income growth, and in doing so, also produces the leakages (savings) from the expenditure stream that provide the room for investment spending, for example.
In this context, fiscal space – which I consider to be the current room that the government has for prudent spending – is defined by the real output gap. A currency-issuing government can always buy what is available for sale in its own currency, including labour that is unemployed.
It can thus spend according to the idle capacity without impinging on the private demand for resources and igniting an inflationary spiral. It is not an exact science because the induced private spending impacts that follow an injection of government spending are variable. But none of that relates to the available savings in the private sector at any point in time.
RK then considered the concept of fiscal space in the Eurozone.
In the eurozone, however, there are numerous government bond markets all using the same currency. There are also free capital flows between these markets. Consequently, countries that would have ample “fiscal space” if they were located outside the eurozone sometimes see their unborrowed savings flee to the government bond markets of other eurozone nations.
A rise in domestic bond yields triggered by capital outflow forces these countries to engage in fiscal consolidation because they no longer have any “fiscal space.”
The problem has nothing to do with where the savings are going. The problem is squarely that the Eurozone governments do not issue their own currency and thus can only get it from taxation or borrowing.
But even within that constraint, if the ECB had played a constructive, rather than its destructive role over the last 5 or so years, there would not have been a bond market crisis.
We know from experience that as soon as the ECB started buying bonds on the secondary markets the yields dropped rapidly and the bond markets were effectively dealt out of the game.
Had the ECB funded fiscal stimulus the Eurozone would have come out of the crisis relatively quickly and the nations would have been able to deploy all the fiscal space available in the form of idle productive capacity and idle labour.
The surrendering of the currency-issuing capacity by Euro nations combined with the recalcitrance of the ECB has caused the crisis. It has nothing to do with what the bond markets did after the crisis emerged.
RK claims that “Spain would have had ample ‘fiscal space’ were it not in eurozone”:
In Spain, for example, private-sector savings are currently running at 10% of GDP, which would be more than enough to finance the nation’s fiscal deficits. This means the country would have ample “fiscal space” were it not a member of the eurozone.
In reality, however, much of this money has flowed overseas, and the fear that the remainder will eventually leave as well has caused Spanish government bond yields to remain at elevated levels relative to yields in Japan, the US, and the UK. The high bond yield, in turn, forced the government into fiscal consolidation with devastating consequences.
You can now all finish of this blog without me using the logic developed previously. Spain would have more capacity to utilise the fiscal space it has – the massive unemployment and idle productive capacity – if it was outside the Eurozone and restored its currency-issuing capacity. That is clear given the ECB’s role.
But it has nothing to with where private savings are or have gone. It relates purely to the fact that it could spend its own currency to bring those wasted resources back into productive use. In doing so, it would create new savings as it stimulated national income.
And if the private bond markets didn’t want to buy any Spanish government debt, the government could just legislate to have the central bank buy it all. Of-course, it could just stop issuing debt altogether (the preferred option) and release all the officials currently engaged in “public sector debt management” into more productive jobs – for example, managing a Job Guarantee program.
RKs work at times reads as if it is core MMT.
But when you appreciate what lies beneath his analysis you realise that like Paul Krugman, he is still operating in a defunct orthodox framework characterised by a failure to realise that the so-called government budget constraint is just an ex post accounting entity rather than a causal, a priori, financial constraint.
That is enough for today!