Stephen Roach spells out five lessons from a decade of quantitative easing policies by central banks

By Stephen Roach*

November 2018 will mark the tenth anniversary of quantitative easing (QE) — undoubtedly the boldest policy experiment in the modern history of central banking.

The only thing comparable to QE was the US Federal Reserve’s anti-inflation campaign of 1979-1980, orchestrated by the Fed’s then-chair, Paul Volcker.

But that earlier effort entailed a major adjustment in interest rates via conventional monetary policy. By contrast, the Fed’s QE balance-sheet adjustments were unconventional and, therefore, untested from the start.

The American Enterprise Institute recently held a symposium to mark this important milestone, featuring QE’s architect, Ben Bernanke. What follows are some comments I offered in an accompanying panel session that focused on lessons learned from QE.

The most important lesson pertains to traction — the link between Fed policy and its congressionally mandated objectives of maximum employment and price stability. On this count, the verdict on QE is mixed: The first tranche (QE1) was very successful in arresting a wrenching financial crisis in 2009. But the subsequent rounds (QE2 and QE3) were far less effective. The Fed mistakenly believed that what worked during the crisis would work equally well afterwards. 

An unprecedentedly weak economic recovery – roughly 2% annual growth over the past nine-plus years, versus a 4% norm in earlier cycles – says otherwise. Whatever the reason for the anemic recovery – a Japanese-like post-crisis balance-sheet recession or a 1930s style liquidity trap – the QE payback was disappointing. From September 2008 to November 2014, successive QE programs added $3.6 trillion to the Fed’s balance sheet, nearly 25% more than the $2.9 trillion expansion of nominal GDP over the same period. A comparable assessment of disappointing interest-rate effects is reflected in recent event studies” research that calls into question the link between QE and ten-year Treasury yields.

A second lesson speaks to addiction – namely, a real economy that became overly reliant on QE’s support of asset markets. The excess liquidity spawned by the Fed’s balance-sheet expansion not only spilled over into equity markets, but also provided support for the bond market. As such, monetary policy, rather than market-based fundamentals, increasingly shaped asset prices.

In an era of weak income growth, QE-induced wealth effects from frothy asset markets provided offsetting support for crisis-battered US consumers. Unfortunately, along with this life support came the pain of withdrawal – not only for asset-dependent consumers and businesses in the United States, but also for foreign economies dependent on capital inflows driven by QE-distorted interest-rate spreads. The taper tantrum of 2013 and the current travails of ArgentinaBrazil, and other emerging economies underscore the contagion of cross-market spillovers arising from the ebb and flow of QE.

A third lesson concerns mounting income inequality. Wealth effects are for the wealthy, whether they are driven by market fundamentals or QE. According to the Congressional Budget Office, virtually all of the growth in pre-tax household income over the QE period (2009 to 2014) occurred in the upper decile of the US income distribution, where the Fed’s own Survey of Consumer Finances indicates that the bulk of equity holdings are concentrated. It is hardly a stretch to conclude that QE exacerbated America’s already severe income disparities.

Fourth, QE blurs the distinction between fiscal and monetary policy. Fed purchases of government securities have tempered market-based discipline of federal spending. This is hardly a big deal when debt-service costs are repressed by persistently low interest rates. But with federal debt held by the public nearly doubling between 2008 and 2017 – from 39% to 76% of GDP – and likely to rise further in the years ahead, what is inconsequential today could take on considerably greater importance in an interest-rate environment that lacks the QE subsidy to Treasury financing.

A fifth lesson pertains to the distinction between tactics and strategy. As lender of last resort, the Fed deserves great credit for stepping into the breach during a wrenching crisis. The problem, of course, is that the Fed also played a key role in condoning the pre-crisis froth that took the system to the brink. This raises a fundamental question: Do we want a reactive central bank that focuses on cleaning up the mess after a crisis erupts, or a pro-active central bank that leans against excesses before they spark crises?

That question – whether to “lean or clean” – has fueled a raging debate in policy and academic circles. It has an important political economy component: Are independent central banks willing to force society to sacrifice growth in order to preserve financial stability? It also bears on the bubble-spotting debate. Yet as difficult as these problems are, they pale in comparison to the foregone output of America’s anemic post-crisis recovery.

That raises two additional questions: Might a pro-active Fed have prevented the crisis from occurring in the first place? And should it be more aggressive in normalizing interest rates?

The Fed’s preference for glacial normalisation both in the early 2000s and now keeps monetary policy on emergency settings long after the emergency has passed. Doing so raises the distinct possibility that the Fed will lack the ammunition it will need to counter the inevitable next recession. And that could well make the lessons noted above all the more problematic for the US economy.

Unsurprisingly, Bernanke offered a very different take on many of these issues at the AEI symposium. He argued that the Fed’s balance-sheet tools are merely extensions of its traditional approach, stressing that “conventional and unconventional monetary policy works through the same channels, with the same mechanism.”

That is debatable. By conflating QE-induced wealth effects with the effects on borrowing costs that arise through conventional channels, Bernanke conveniently sweeps aside most of the risks described above – especially those pertaining to asset bubbles and excess leverage.

Ten-year anniversaries are an opportunity for reflection and accountability. We can only hope that circumstances don’t require another unconventional policy experiment such as QE. But in the event of another crisis, it would pay to be especially mindful of QE’s shortcomings. Unlike Bernanke, I fear there is good reason to worry that the next experiment may not work out nearly as well.

Stephen S. Roach, a faculty member at Yale University and former Chairman of Morgan Stanley Asia, is the author of Unbalanced: The Codependency of America and China. Copyright: Project Syndicate, 2018, published here with permission.

We welcome your help to improve our coverage of this issue. Any examples or experiences to relate? Any links to other news, data or research to shed more light on this? Any insight or views on what might happen next or what should happen next? Any errors to correct?

We welcome your comments below. If you are not already registered, please register to comment or click on the "Register" link below a comment.

Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current Comment policy is here.


One of the more thoughtful..

But still an economist - still fails to see the finite nature of the underwriting planet and the inevitable cessation of compound growth. How many times has this got to be pointed out?

Daly calls it the fallacy of misplaced concreteness:

Surely growth will almost certainly continue while technology advances.
With improved technology we may not need to deplete finite resources - for example the sun beams down plenty of energy on a daily basis. I guess you could say the sun is finite, depending on your time frames...

JJ - we've probably moved on from there.

You can't do copper with solar :) But growth is exponential (or wants to be) and technology can only give us efficiencies - it can't create energy. PV is fairly near it's practical limit, so too are electric motors -it becomes a graph of diminishing returns, laid over on of diminishing supplies.

Don't get me wrong, it's the way to go - it just won't support growth as we temporarily knew it.

I'm sure if we covered every desert in the world with solar panels we would have plenty of energy. The technology exists, we just don't use it yet because we still have the finite resources available.
And the efficiencies can be pretty impressive - compared to 10 years ago my light bulbs use 90% less energy, my phone uses a fraction of the power that my old PC did (while having a lot more processing power), my heatpump uses a quarter of the energy of my old heaters, and my car uses about half the petrol of my old car.

Will those efficiencies allow us to double the production of those items (and a billion other products) without further degrading the environment/natural ecosystems, and decrease the cost of living, debt levels, inequality, increase the quality of living for every individual and allow every individual both current and future to maximise the accumulation of material "wealth" indefinitely?

That depends on whether we can keep making efficiencies forever. I'm sure there are a lot more we can make. We will no doubt make some big efficiencies in food soon - moving to a much higher plant based diet.

Inequality is a different issue and in my opinion much more important and difficult to resolve.

Efficiencies don't matter if the narrative doesn't change. Jevon's parradox!

The study of economics should include the sciences. "I don't understand it, so I won't measure it".

Yeah, but if that piece represents "more thoughtful" (i.e., more robust than most) academic analyses - we are in big trouble - as even this author clings onto the notion that this debt will someday be repaid - and that there will be a "next time" under the current monetary policy paradigm. His last sentence communicates his application of this false assumption;

Unlike Bernanke, I fear there is good reason to worry that the next experiment may not work out nearly as well.

What he fails to acknowledge is that the experiment made the patient terminal (i.e., the global monetary framework is a dead man walking).

Also, the author explains;

That question – whether to “lean or clean” – has fueled a raging debate in policy and academic circles. It has an important political economy component:

The linked academic paper gets closer to that admission;

There is a need for a new macrofinancial stability framework that would use both regulatory and monetary instruments to resist credit bubbles and thus promote sustainable economic growth over time.

But not quite willing to just plain state that the old paradigm clearly and simply failed.

To assume that the "patient is terminal" and the end is nigh is essentially the same as predicting the top/bottom of a market. In all likelihood things will carry on a lot longer yet, complete with the usual ups and downs. So can we expect another experiment down the road? I'd say yes.

It's only the end of this monetary (means of exchange) paradigm that I think is nigh. Logically, it must be superseded by another as this level of global debt can never be paid.


I hope you don't mind me interjecting on your post, but some commentators may not understand the bigger picture on economic cycles. Several minor recessions are generally interspaced every lifetime with a big one, which is either very deep or very long. No one believes it can happen again, because often the previous generation to experience the last 'biggie' have all died. But, they have been prevalent throughout history and require a major re-set of the over-exuberance (major deflation) to re-set. Sadly they often also have ended up in War, which has sped up creative destruction and provided a levelling of things to induce the next demand and growth cycle.

Exuberance of late 1700's was mercantilism.
Exuberance of mid 1800's was steam, power and locomotion
Exuberance of 1920's was mechanisation and automobiles
Exuberance that we have been trying to avoid the effect of since the 2000's is information technology and the internet revolution, but where valuations are still wildly above the returns to capital - and this is the case with many of the massively valued stocks are seeing today. The link below highlights the likelihood of a big correction to take place still, QE, in my opinion has merely postponed its arrival, but the super cycle cannot be denied it pound of flesh payment. The 'Great Recession' was hardly that but the media is very dramatic these days with its labelling of events.

If you want to go back further you will get to Tulips in the 1630's... Generally every 70-80 years there is a biggie. The question is did we really have it already? Was that it?

And for the specuvestors....... I know New Zealand is diffrunt.


History will remember QE as a drug that saved a dying patient , kept the patient alive on life support and finally given a euphoric stimulus taking the patient on a unsustainable false reality. Perhaps it might have been better for nature to take its natural course. Only time will tell.....

Yep. When Bear Sterns went down, the Government should have let the carnage in the merchant banking sector play out, given the Fed is the merchant banking sector, it was an entirely compromised institution.

Any discussion about economics, or politics, that does not discuss resources is incomplete. Half arsed you could say, perhaps even meaningless. The question is whether QE was linked to an increased consumption of resources, which much happen according to the current narrative, or prevent the contraction of such consumption. The argument is perhaps that it did both.

The next question is did it change the underlying trend in the consumption of resources, and the most useful question might be what is that trend? Can that trend be altered in any meaningful way by central bank policy?

What is the ratio of money creation to the consumption of resources? Is this useful?

I believe there is a growing disparity between the rate of credit creation and the rate of consumption of resources, the differential growing exponentially. Can these two ever be reconciled? I would propose it has to be, and so further propose the question, what does that process look like?

I've wondered a similar thing too. Quantitative easing appears to have driven up asset and shares prices but I'd be surprised if they have led to increased overall resource consumption. For example, the high cost of accommodation in NZ now means a significant amount of household income (particularly in the younger cohorts) is spent on this debt payments or rent rather than purchasing consumer items or traveling (resources).

I'd guess QE was/is linked to increased consumption of resources. Even looking at NZ since 2008, it seems we have accelerated the depletion of aquifers and increased pollutants into our waterways by industry credit expansion.

As I have been saying for years,10 years of QE has only very marginally boosted the economy, but produced a range of very strong and undesirable side effects. The link between QE and boosting the economy is very weak, but it is very strongly linked to inappropriate wealth redistribution and creating asset bubbles. This ultimately is likely to lead to future economic crises. Long term probably very counter-productive and we desperately need a long hard think about how to manage our economy and banking sector.

The proposition that we need positive inflation is completely contradictory to the functioning of a healthy competitive economy, in which competition, innovation and technology will increase productivity and drive down prices. The "imperative" for positive inflation means that in the face of increasing productivity driven lower prices in some sectors, cash is pumped into the economy driving up prices in less competitive sectors and fuelling asset bubbles. From my limited understanding the only three reasons that I can see for the need for positive inflation are
1 That the banking sector is very dependent on it to make their extortionate profits. Remember my earlier posts where I revealed that the NZ finance sector generates as much profit as the other top 200 companies in NZ and most of that is accrued in the 4 big Australian banks.
2 We prefer to mange our economy with the belief that future inflation will pay off today's capital expenditure. This is really lazy thinking and a self deception. If we took a more honest and disciplined approach we might have an economy and finance sectors that are a lot less sexy and volatile, but far more stable and sound.
3. It suits the wealthy people with lots of political influence because it makes it easy for them to borrow money and invest in assets of all forms, knowing that inflation will pay down the debt. - Farming inflation.

An altogether more pointed, insightful analyses than that of the author of this piece.

The good Dr Roach simply told us what seems to have been bleedingly obvious to the everyday man/woman on the street.


Excellent Piece.. Basically, Beveridge economics doesn't stand up to scrutiny when faced with the realities of the 70-80 year economic super-cycle (a deflationary bust)... It was enacted on the assumption that the Great Depression had been a 'one off event and would never happen again. QE has delayed the onset but sadly has raised the height from where the fall starts. which may make it more painful at landing..

Bit of protectionism anyone? How about a few tariffs. Is history repeating?

Happy 10 years of stupid!

Its been a wild illusion of prosperity and wealth!

Of course the Elephant in the room is the $100 Trillion dollars of current global debt, or approx $13200 for every man, woman and child alive today and growing by the second

Many Happy Returns!

Happy Birthday!

The point is QE is a poor substitute for fiscal policy. It is an asset swap. Central bank swaps cash (reserves) for bonds. Might have some effects on interest rates at the margin. But doesn't increase lending in real economy for investment to offset balance sheet recession and a lack of demand. Now, if the austerity binge hadn't happened things would have been different. Time to brush off the dust on Keynes's General Theory and learn about its modern successor - MMT. Monetary policy is not nearly as useful as they'd have you believe. We need are much strong automatic stablisers - like a job guarantee. Dry economists like Larry Summers and Olivier Blanchard even say as much. And the all acknowledge that the idea of expansionary fiscal contraction - aka austerity - doesn't work. Positive fiscal multipliers do exist. And that QE is not much use.