This submission makes the case for a “balance sheet approach” to QE that distinguishes QE for consumption from QE for investment and seeks to use QE to raise national net worth. There are an investment and productivity gaps in Britain that may be caused by market failures and economic structure. QE might be adapted from its role in short-term monetary policy to finance productivity-enhancing investment.  “Productive QE” will not raise inflationary expectations nor the prospects of currency depreciation. To preserve the Bank’s independence and mitigate political problems, investment decisions should be located within a strategic investment fund.  The Bank can initially buy this fund’s securities and subsequently sell them in financial markets once projects have come to fruition.

  1. This submission seeks to address questions 7 and 11, respectively:
    1. “What trade-offs does the Bank of England’s Quantitative Easing programme entail? What effect might it have on inflation?”
    2. “What evidence is there for any upper limits to the Bank of England’s Quantitative Easing programme?”
  2. A broad analysis will inevitably overlap with some of the inquiry’s other questions, especially questions 1 and 10 that concern the Bank’s independence and QE’s impact on economic stability, respectively.
  3. The submission is made in a personal capacity and the contents have not previously been published. The perspective is that of an entrepreneur and investor.
  4. The motivation is to challenge a nascent orthodoxy of “QE infinity”. The ideas set out here are that some QE can be useful but we need to understand the uses that are made of new money. This presupposes a nexus between monetary policy and fiscal policy. We need to add a more “granular” approach to this way of thinking, one that considers the structure of the economy as a whole and makes room for a distinction between consumption and investment.
  5. As perceived today, the QE programme implies trade-offs for the value of the pound, for the efficient allocation of assets, for the difficulty in accessing housing and pension assets, and for financial stability. Each of these matters in its own right. In particular, some commentators, academics and practitioners have a misapprehension about the currency and a bias that favours “talking the pound down”. I believe that this is wrong. Given the right conditions, QE need not imply trade-offs for inflation (question 7).
  6. Depending on how QE is used, and how much, there need not be any fear of a permanent rise in households’ expectations of inflation (question 11).  Where this limit lies should be driven by some elements of classical finance theory:
    1. The uses made of new money;
    2. Whether QE facilitates the creation of new assets;
    3. The rates of return generated by those assets, and
    4. The likely costs of financing those assets.
  7. This is a “balance sheet approach” to money creation that recognises whether an asset is created or whether funds are spent, and the overall effect on national net worth.
  8. The central argument is that new liabilities of the Bank of England are applied to productive activity that creates assets for the nation, increasing national net worth.  Increasing net worth raises prosperity and well-being.
  9. In an economic sense, and a modern accounting sense, assets are activities that generate a return above their cost, including their cost of capital. In finance theory, they have positive “net present value”. By definition, qualifying assets raise productivity, such as comprehensive broadband roll-out, investment in human capital, e.g., vocational training, or the creation of new laboratory or research facilities. They can include “moonshot” assets that yield technological returns over a distant horizon. The classic example is U.S. defence research in the 1960s that generated technologies that were later combined to enable the Internet.
  10. Their value in a national accounting context should be measured by their future revenue-generating capacity, as new job-creating activities and by their potential to add to national economic productivity and export competitiveness.
  11. The market will perceive the creation of valuable assets (which implies raising net worth) as a “good” use of QE – the Bank’s new liabilities do not lead to higher inflation.  Instead, new money creation sustains investment that will raise productivity and export competitiveness in the long run.
  12. At present, QE is perceived as being used to sustain consumption while large parts of the economy are in furlough. Sustaining consumption once the economy is released from lockdown will be inflationary. Equally, devoting QE funds to investment that would happen anyway, say to building more houses, is also likely to be inflationary as land prices, building materials and labour costs are all bid up.
  13. I have described a balance sheet approach to QE that seeks to use new Bank money for investment that would not otherwise take place. The overall effect is to raise net worth by creating new assets. The plausible limits to this activity are harder to ascertain but the analysis gains from looking at levels of investment made by other advanced economies.
  14. For the past twenty years, the sum of investments in fixed capital made by the government and the corporate sector has held steady at twelve to fourteen per cent of GDP per annum (Figure 1 in appendix). Since at least 2000, this level is consistently at the bottom of a league table of industrialised economies that includes Germany, Japan, Sweden, Switzerland and the United States. At the top end, the Swiss and the Swedes typically invest between twenty and twenty-two per cent of GDP annually. The data are from the OECD and should be comparable across countries.
  15. Business is responsible for the bulk of this investment (Figure 2) and the corporate investment share of Britain’s economy is also relatively low.  Differences may be explained by economic structure (manufacturing versus services, or the importance of exports) and the types of investment required to sustain a particular economic structure (in acquiring land over spending on human capital, for example). However, it is worth noting that investment in intellectual property (IP), which should lead productivity growth and competitiveness, also illustrates the existence of a “gap”. Britain typically invests three to four per cent of GDP in IP, a level that is comparable with Germany’s (Figure 3).  The Swiss, whose exports face serious currency headwinds (in the form of appreciation), need to differentiate their export goods. They invest seven to eight per cent of GDP in IP, and the rate of investment has grown consistently since 2011.
  16. There is, potentially, a gap of seven to eight per cent of GDP per annum between Britain’s level of investment and that of some of the world’s most advanced and most productive economies. There will be complex explanations that include differences in institutions, education, regulation and incentives. These are beyond the scope of this inquiry.
  17. If future QE can be used to plug this gap, then there may be an absolute, upper limit to non-inflationary QE equivalent to about eight per cent of GDP per annum. The proportion of useful investment that will not be met by the private sector may set the plausible ceiling at some lower level. Market failures will account for the difference. For the sake of argument, if half the investment will not take place then non-inflationary QE may have a ceiling of about four per cent of GDP, or £80 billion per annum. If the balance sheet approach is correct and non-inflationary QE is subject to a “positive net worth constraint” – then the investment gap may help to identify the limiting parameters of QE.  More analytical work is needed.
  18. With the exception of Japan, productivity growth in Britain as measured by GDP per hour worked has also lagged (Figure 4) since 2007. The Americans have managed about one per cent annual growth, while the Swiss and the Swedes are around 0.7 per cent. Britain is on 0.2 per cent. These small differences accumulate over time and lead to widening divergence that affects standards of living: the productivity gap with the United States has widened by 11 per cent since 2007.
  19. The underlying assumption is that investment, and particularly business investment, has a positive effect on productivity growth. The next step is to argue that plugging the “investment gap” might help the productivity gap.  This is a complicated story but there is some correlation between the two (Figure 5), although detailed argument about this important topic is beyond the inquiry’s scope.
  20. The balance sheet approach brings economic structure and investment into the analysis. Monetary economics seems uncomfortable analysing national economic characteristics: two of the leading contemporary textbooks on the subject devote only a handful of pages to investment.[1] The idea is that we need to draw starker lines in monetary and fiscal policy-making between consumption and investment, and between different kinds of investment, placing a higher value on what is most enhancing of national productivity.  It matters where QE “goes”.
  21. For now, the Bank buys Gilts blindly, because it is not mandated to exercise judgment about the causes and uses of public-sector deficits. This may be acceptable for lots of reasons during lockdown, notably the support for furloughed jobs. At some point this will end and the Bank’s continued purchase of Gilts without limit makes it a party to activist fiscal policy unless it can be certain that, first, the Gilts will retain their value and, second, that there will be future market demand to absorb these Gilts.  Whether either of these will be true cannot be known today.
  22. Two problems arise. The first is that we have become used to thinking of monetary policy as a tool to manage short-term fluctuations, but now QE has been with us for the long run without apparently touching the supply side or long-run productivity growth.
  23. The second problem is that QE buys existing assets, most recently “very lightly used” assets, or freshly-issued Gilts. Generally though, the policy is a transfer of funds from buyers – in this case the Bank – to sellers. Surely many of the sellers of Gilts are recycling their new-found funds elsewhere in credit and equity markets. It is now a commonplace to say that QE has driven up prices (but perhaps not fundamental values) in these markets which makes pension assets both more expensive and riskier – are prices of Gilts and shares more likely to rise or fall from existing levels? We do not know, but the risks are now heavily skewed in one direction. If productivity levels have not changed, then the fundamental values of these assets are unlikely to have changed much. But their prices have on the fiction that the “discount rate is lower”.
  24. There may be an alternative use for QE that addresses these problems. The question that we need to ask is, “how can the Bank best support the real economy through its QE programme?” To be sure of retaining its independence the Bank might consider restricting securities it buys to those that finance the creation of valuable assets (rather than monetary financing). If the investment and productivity gaps described in paragraphs 14 to 19 can be financed with dedicated securities, the Bank would undertake to buy some proportion (or all) of this paper. The securities are in effect claims on infrastructure, research or educational projects that can be evaluated by an independent agency. The Bank would want to assess the value of this agency’s assessments, and do so independently. Market failures mean that financial markets cannot value the claims on uncertain returns, so will not buy associated securities.
  25. This discussion leads to the idea of some kind of strategic investment fund.
  26. Suffice it to say that for such an idea to be credible, investment decisions need to be separated from the political process and from the Bank. Pork barrel investments or any state-led attempt at “picking winners” will be of no use. These efforts do not usually end up financing the creation of sustainable and high-quality jobs. Equally, the Bank is not best placed to prescribe investment allocations. There is space for an autonomous agency to issue the relevant securities and make the relevant investment decisions in productivity-enhancing assets.
  27. Perhaps we can call this “productive QE”. It would limit the risks of both higher inflation and currency depreciation. Mitigating the inflationary (consumption) perceptions of QE will lead to greater currency stability and improve exporters’ ability to plan and give them more incentives to invest.
  28. Under this scenario, the Bank’s new liabilities are not recycled into financial markets. We have mitigated the problem of pumping up asset prices and the risk of inflationary monetary financing.
  29. There are underlying hypotheses that, first, the investment gap should be plugged and, second, that private-sector funding will not be available because of market failures. Relevant projects may require longer horizons than private markets usually endure, have return profiles that are too uncertain and do not allow for considered valuation, or offer collective benefits that cannot be claimed by any single entity. Only a deeper assessment missed opportunities in Britain’s economic structure can ascertain whether these hypotheses are plausible.
  30. If the hypotheses are correct, projects come to fruition and market failures correct, allowing the private sector to value both the projects and their issues of paper. These might then be sold to the private sector providing a mechanism for the unwinding productive QE.
  31. The other side of this argument is that, if money creation persists and fuels a permanent increase in consumption demand that strains productive capacity, we can expect higher inflation and the pound to weaken. A weaker currency does not systematically imply an improvement to the balance of trade: we would probably import more goods (and inflation) than we would export.
  32. Sadly, there is much nonsense said about the merits of currency depreciation. A weakening of the currency reliably benefits exports only to the extent that these exports are commoditised, i.e., that they are sold in competitive markets largely on the basis of price, rather than quality.  Assumptions that the United Kingdom will become more competitive because the pound is weaker are misguided unless the country predominantly exports commoditised goods. For Britain to compete this way means downward pressure on wages which is hardly consistent with a “levelled-up” economy.  Instead, we should strive to compete on quality and differentiated goods and services in a way that renders currency fluctuations largely irrelevant.  Switzerland remains a competitive exporter of high-quality goods despite persistent (and sometimes extreme) trade-weighted appreciation of the franc. This state of affairs endured throughout the inflationary post-war decades and the Swiss have had to learn to adapt accordingly. Much has been said about the merits of Swiss research and technology institutes, and the cooperation between industry and vocational training establishments.
  33. A weaker currency does not guarantee higher exports, as Sweden found after its own banking crisis in the early 1990s and after the financial crisis of 2007-2009. A bias towards currency depreciation and higher inflation relative to our neighbours does not ensure prosperity in the long term.
  34. Both real and nominal rates of interest have fallen far since 2007 without, apparently, any increase in investment (Figure 1). This is contrary to received monetary thinking and may say more about the state of economic theory than about policy choice. Perhaps there is a counterfactual argument that investment would have fallen without the fall in rates. The data from other countries do not seem to bear that out.  It appears more likely that fundamental differences in economic structure explain persistent differences in national levels of investment.
  35. Short-term and long-term real rates of interest are negative. For now, domestic savings may be high and excess savings will flow whither they are most easily absorbed. It is much less bother to invest in the stock market than to start a new business, build a plant or train a biochemist. This implies continuing “financialisation” of the economy: resources applied to financial trading that shifts money from the buyer to the seller and back, rather than creating new assets.
  36. The expectation that real interest rates will remain low or negative sows the seeds of future financial instability. This is a point made by Claudio Borio, the chief economist of the Bank for International Settlements, and others at the BIS.[2] As lower rates raise both asset prices and expectations of future asset price growth, demand for credit linked to such assets – housing – acts as an “accelerator”. Asset price bubbles become more likely.  In our recent history, when a bubble has burst, monetary policy has responded with lower interest rates, fuelling a long trend of ever-decreasing rates and pumping up asset prices, that raises the risk of market instability and bank credit losses. There are limits to how much more loss-absorbing capital banks can raise without a radical overhaul of their business models.  As they stand now, bank profits largely fail to meet their costs of capital.  This has other implications for the future of the banking system.
  37. The right uses of QE need not condemn us to having to make these trade-offs, but we need to change how we do some things. British policy-making is not particularly good at distinguishing between spending and investment, or at the evaluation of investments. We too often talk about “investing” when we really mean spending. This may be politically expedient but it creates bad habits and prevents households from understanding the trade-offs, and the consequences for their future prosperity.
  38. If we can communicate, to households, businesses and financial markets, that QE is being directed at raising net worth, then it should follow that the spectre of higher inflation will not appear. Confidence in Gilt markets can be sustained this way, as well as in the Bank’s policy-making and its independence. The Bank would not be perceived as financing the government’s deficits without end or without question.
  39. The principal task ahead, therefore, is a deeper analysis of the causes of Britain’s limited investment and an assessment of the scale of opportunities to plug the gap. Favourable findings can lead to productive QE in the presence of suitable governance arrangements that do not politicise investment decision-making and do not implicate the Bank in the assessment of particular projects.
  40. Markets may become concerned to the extent that QE-financing of projects crowds out private-sector access to financing. However, these projects are not undertaken precisely because of market failures, making any concerns about crowding out the private sector irrelevant.
  41. In summary, this submission makes the case for a “balance sheet approach” to motivate “productive QE”. Quantitative easing can be applied to creating new, productive assets that raise national net worth and prosperity. Such projects will be the fruit of identifying the causes of an investment gap that has long persisted between Britain and other advanced economies. The assumption is that market failures mean that such projects are not financed by private means. Productive QE should mitigate the risks of higher inflation and currency depreciation. There may be an argument in favour of a strategic investment fund.

9 February 2021



Figure 1:

Sum of corporate and general government investment as % of GDP

Sources: OECD data; author’s calculations.

Figure 2:
Corporate investment in gross capital formation as % of GDP

Sources: OECD data; author’s calculations.

Figure 3:
Investment in intellectual property as % of GDP

Sources: OECD data; author’s calculations.

Figure 4:
Productivity: GDP per hour worked

Sources: OECD data; author’s calculations.



Figure 5:
Productivity growth and corporate investment

Sources: OECD data; author’s calculations.


[1] See, for instance, Walsh, C.E. (2017) Monetary Theory and Policy, 4th ed., MIT Press, or Mishkin, F.S. (2016) The Economics of Money, Banking and Financial Markets, 11th ed., Pearson Education.

[2] Borio, C., Disyatat, P. and Rungcharoenkitkul, P. (2019).  “What anchors for the natural rate of interest?”  BIS Working Papers No.777.  Bank for International Settlements, March.