Everyone has heard of quantitative easing, but how many people actually understand how it works? Most people vaguely understand that it involved the Bank of England creating new money, but describe this in terms of “printing money”, which is so different to the actual process that it misleads you right from the start.
And it’s amazing the extent to which even those who do understand the process get important details wrong. While looking for references for the blog and fact-checking a couple of points, I consistently came across commentators (economists and journalists) who had 80% of the story right, and then allowed a misunderstanding of a nuanced point to magnify into a serious error. So this week I’m going to set out how quantitative easing (QE) works in precise detail, and then explain how this confirms everything I’ve been saying about financial markets.
How QE Works (UK version)
In sum, with QE the central bank simultaneously creates new ‘broad money’ (the type of money we all spend) and new central bank reserves, and uses it to buy government bonds (i.e. government debt) – a very small amount of private bonds has also been purchased.
That’s it in a nutshell, but the details matter. In the UK, the Bank of England owns an “Asset Purchase Facility” (APF). This is a wholly owned subsidiary of the BoE for the purpose of buying assets (as the name implies). With QE, the BoE loaned the money to purchase government bonds to the APF. This loan created new broad money, just as new money is created when a private bank makes a loan (as you should know by now).
The APF bought government bonds off the market through an auction, typically from insurance companies and pension funds (ICPFs). When the purchase was made, the new money with the APF had to be deposited in the bank accounts of the ICPFs. This meant that the Bank of England had to transfer central bank reserves to the reserve accounts of the corresponding banks. This central bank reserve was newly created, at the touch of a button on a computer. So here we see the creation of both broad and narrow money – let’s explore what happened to both types of money.
New Broad Money “Injected” into the Economy
The ICPFs now had deposits from the sale of government bonds sitting idly in the bank, so they bought more financial assets off the private sector in the search for greater returns. This demand for assets pushed up asset prices, which is exactly what the Bank of England wanted to happen. Higher asset prices means that debt is cheaper for those selling assets, and the BoE expected that this would encourage businesses to take out debt and fund investment.
This is the primary way that QE was intended to work. “Monetary transmission” was intended to flow from the ICPFs into private asset markets and into the real economy. QE did not fund public spending by printing money (but it did effectively reduce the national debt, a point we will come back to later), nor did it “give free money” to private banks to lend (discussed in the next section).
This is described in detail in this paper on the subject by the BoE in 2009, and a very clear summary is given on pages 21 and 24-25 of this paper on money creation in 2014 (one I’ve often cited in the past), particularly dispelling misconceptions that had arisen in the intervening years.
New Central Bank Reserves and Bank Lending
In addition to this process, QE massively increased the amount of central bank reserves in the banking system. You will often see people saying that banks “lent out” the new reserves, and others claiming that this is nonsense because banks cannot lend out reserves. The latter view is correct – “nonsense” is perhaps a bit harsh because QE did enable banks to start lending again, but it is important to be clear on how this works.
Central bank reserve (or ‘narrow money’) is the money that banks hold in reserve accounts with the central bank. Every time we make payments to customers of other banks (e.g. cheques or debit cards) central bank reserve is transferred between the reserve accounts of the corresponding banks. If a bank lends out more in a day than it receives in deposits, it will have a shortfall and will need to borrow in the interbank overnight lending market. If other banks believe a bank is in trouble they won’t be willing to lend to it and it could go bust (which is what happened to Northern Rock).
QE flooded the market with reserves, and took away the risk that banks could run out. This made banks happy to lend again, and helped us work our way out of the credit crunch. This was also a stated purpose of QE, but not the primary one (stated on page 93 of the 2009 paper).
Impact of QE
Both processes described above were seen by the BoE as ways of “injecting money into the economy”. In fact, in their 2009 paper they talk about “injecting” 22 times – they clearly wanted to convey an image that this was like giving the economy an adrenaline shot. So what was the result?
As you would expect, there have been many studies on the impact of QE with a bewildering variety of conclusions, but I can do little better than point you to this BBC Radio 4 programme on the subject (much as I would rather be able to point you to a paper with easily accessible references to check the facts).
The programme outlines clearly how and why QE was meant to work (including the point that the new reserves were created at the touch of a button). It gives a couple of examples of where the sale of corporate bonds did enable investment that created new jobs.
But the programme makers researched the sale of corporate bonds following QE, specifically which sectors these came from (this is where I wish they provided a link to the results of this research). They found that the vast majority of such bond sales were from the financial sector. It was financial companies selling bonds to finance their own lending. An interviewee from the credit ratings agency Moody’s estimates that only 0.8% of the money created via QE actually found its way to investment to increase productivity. Not much of an “injection” after all, then!
But what about the lending by the financial sector? Where did this end up? Small and medium size enterprises (SMEs) tend not to have access to bond markets. They rely on borrowing from banks. As explained earlier, QE helped ease the credit crunch so that banks were ready to lend again. But this lending has gone on their favourite source – mortgages.
So the new money created by QE flowed into financial and property markets. The increased demand has pushed prices up. In other words, QE simply reinflated the asset price bubble, the bursting of which caused the last financial crisis.
And the biggest tragedy here is that increasing the money supply, as QE did, was exactly the policy that was needed, but the newly created money needed to be directed into productive investment. This policy can still be implemented, but of course the room in the system for new reserves has been reduced by the failed attempt at QE. It should be pointed out that the implementation of such a policy fell outside the remit of the BoE – it would have needed Government action backed by Parliamentary approval. The extent to which the BoE can be held culpable for the failure of QE therefore lies primarily on whether it advised the Government that more was needed, and the internal conversations between the Governor of the BoE and the Chancellor are not a matter of public record.
The incoherent application of QE is even more of a missed opportunity when you understand that the policy effectively paid off one third of the national debt.
QE and the National Debt
As a result of QE, the BoE now owns about a third of UK government bonds. In other words, the UK Government owes that money to the BoE. But the UK Government owns the BoE, so it owes the money to itself. And you can’t owe money to yourself, so in practical terms that debt has been paid off.
Now, this is something of a simplification. The BoE could sell those bonds back into the market, and has always talked in terms of someday “reversing QE”. Suddenly the debt would reappear, even though the Government wouldn’t have taken more debt out and benefited from the proceeds. However, prior to Brexit the only conceivable reason the BoE would do this is if the economy is booming and inflation needed to be controlled. If we have a booming economy we don’t need to worry about higher debt. Since the Brexit vote we’ve seen inflation without growth, and Brexit itself carries a high risk of stagflation, but who knows what will happen and how the BoE will respond – it would take far too long for me to discuss all these possibilities now.
In any case, the new international banking regulations of Basel III require banks to hold much higher levels of high-quality liquid assets (HQLA) than they did before – basically, banks need to have sufficient liquid assets to be able to keep going for 30 days in the event of another financial crisis and credit crunch. And the main form of HQLA is central bank reserve. (Don’t confuse this with reserve requirements, which are irrelevant and don’t even exist in the UK.) Pozsar has an excellent graph on page 11 of this paper for Credit Suisse titled “What Excess Reserves?”, showing that we could in fact think of the very low level of reserves before as being strange, and the high levels we see today as “normal”. Under the new Basel III regime there is every possibility that QE will never be reversed.
Another correct point you can find online is that the public sector has swapped one liability (government bonds) for another (central bank reserves). Prior to QE, government bonds were a liability on the balance sheet of the government. After QE, the relevant bonds are still on the government’s balance sheet as a liability, but also on the BoE’s balance sheet as an asset – so on consolidated accounts for the entire public sector these balance out. However, the newly created reserves are also a liability for the Central Bank, so QE has not reduced the total public sector liabilities at all, it’s just swapped one liability for another.
On the rare occasions you find someone making this point, they are correct. However, usually they will then say that it is therefore nonsense to say that Government debt has been paid off. And this statement is so lacking in nuance as to be nonsense itself. When people think of debt they think of something that you have to repay. In accounting terms, central bank reserves are a liability of the central bank and therefore of its owner, the UK Government. However, this liability does not require a “repayment”. The economic implications of this liability are so different to the implications of privately held government bonds that it is misleading to say that increased reserves is the same as public debt. Central bank reserves need to continually increase in line with productivity, and as mentioned above the new regulations of Basel III need a far higher level of reserves in the system to make it work.
So we come back to the fact that about a third of UK national debt is owed to the BoE. The Government pays no interest on this debt – the debt is free. It would be simpler and more accurate to report a new figure for UK national debt that is a third lower, but the UK Government always cites the higher figure. There are several possible reasons for this, but I just want to highlight two to give different aspects.
I believe that the BoE wanted to make this radical policy seem as “normal” as possible, and therefore would talk up the future reversal of QE. By way of contrast, in his recent papers for Credit Suisse, Pozsar is continuously stating that high reserves is the new normal, and we need to get used to it.
As for the Government, there are many who would suggest it wants to justify its policy to reduce the size of the public sector so it continues to cite the inflated figure of national debt. I have sympathy for this point of view, but it would not apply to the Labour Government that first implemented QE. In that case, it may well be that they wanted to avoid a public narrative that they were “printing money” to pay off the national debt.
And in both cases, never underestimate the extent to which politicians simply do not understand any of this at all. Therefore, theories on the reasons for the actions of politicians that assume they do understand it are probably inaccurate.
Conclusion
The creation of new money to relieve the credit crunch was necessary, but the specific way in which QE was implemented was a tragic missed opportunity. The lessons from QE confirm the conclusions drawn on this blog about financial markets in general.
First of all, I concluded in the section on ‘Money‘ the need for a proportion of the Government deficit to be funded each year by the creation of new money, with the exact amount to be calculated according to economic criteria, not the desire of the Government to spend. (When the economy is over-heating this would not apply and a Government surplus may be necessary.) QE did indeed repay one third of the national debt, and could in fact have been used to fund investment through public infrastructure projects as a more effective means of “monetary transmission”.
In the section on ‘Productivity‘ I concluded that if individual acts of saving are not channelled through to investment they will simply inflate asset prices, while the underlying value of the economy stagnates through lack of investment. This is exactly the condition of our economy right now, and we can see that QE has been one cause of inflated asset prices. Indeed, the very purpose of QE was to increase asset prices. It may seem insane when I put it like that, but the purpose was to make borrowing cheaper for businesses. As we have seen, the policy did not in fact stimulate greater investment to any significant degree, and active steps were needed if this were to happen (steps that were outside the powers of the Bank of England). Which leads to…
The current section of the blog, ‘Financial Markets‘, has looked at the seismic shifts in financial markets in this century – a complete transformation that turns on its head the traditional model described in textbooks. We have seen that these shifts are driven by global imbalances (e.g. in wealth and in trade) that have destabilised the economy, and noted Pozsasr’s conclusions that if we want a stable economy that is not misallocating capital resources and thereby leading to bubbles and crashes we need to address these imbalances.
Finally, I reached the conclusion in the most recent post that: “…to create an economy that provides enough for everyone we also need to reform financial markets to ensure they actually channel funding to productive investment.”
Similarly, the lesson we learn from the failure of QE is that steps needed to be taken to ensure that the newly created money was channelled to productive investment. There is much we can learn from this failure, and I hope that spending so much time this week on this analysis has made clear the critical importance of such steps.
In two weeks I will conclude this section of the blog on the need to reform financial markets, but before that I will look at this subject from yet another angle – William Lazonick’s brilliant empirical work on the the buyback of shares by large corporations. That may not mean anything to you, but believe me you don’t want to miss it.
Postscript:
The term “quantitative easing” was actually coined by the German economist Richard Werner, who worked for many years in Japan during its long period of economic stagnation in the ’90s, known as the ‘lost decade’. In 1994 Werner proposed a policy of creating reserves to stimulate credit for real economy transactions (as opposed to credit for financial speculation). The term he used in Japanese translates directly as “quantitative easing”, and the purpose of the new term was to distinguish it from traditional monetary policies, such as simply increasing reserves without then targeting new credit to the real economy.
His policy was derided by both Keynesians and monetarists, but the label caught on. Somehow its meaning got twisted simply to mean the creation of new central bank reserves, but without the necessary targeting. This was exactly the policy that Werner had said would not work. In 2001 the Bank of Japan adopted just such a policy, using the name quantitative easing (in Japanese). The policy failed, just as Werner predicted, and was abandoned in 2006. He tells the full story of this here, and his theories are brilliantly explained in his book “New Paradigm in Macroeconomics“.
After the great financial crisis, QE was adopted by central banks around the world (the Bank of Japan version, not the Richard Werner version). The creation of new reserves was absolutely needed, but the resultant credit creation had to be targeted at the real economy, just as Werner advocated in Japan in the ’90s. The failure to learn from Japan’s mistakes led to the missed opportunity of QE, as described in this post.
