We live in a world of low interest rates and volatile asset values. Over the last three decades, real interest rates have declined continuously throughout the western world, having been in negative territory throughout much of the last decade. Although there are different views as to the ultimate cause of this decline, most agree that it reflects some form of asset scarcity, i.e. an increased demand for assets that raises their price thereby depressing returns.
Over these same three decades, Japan, the US, and parts of the euro area have all exhibited large booms and busts in asset prices with significant consequences for economic activity. The long slump that has characterised Japan since the early 1990s is commonly interpreted as the result of the collapse of real estate and equity bubbles. In the US, the last recession has also been largely attributed to the development and subsequent burst of a large bubble in the real estate and equity markets. In the same vein, some of the most severe recessions experienced in the euro area in the aftermath of the global financial crisis – such as those in Ireland and Spain – have coincided with the bursting of real estate bubbles.
How should monetary policy be conducted in this ‘bubbly’ world of low interest rates and large booms and busts in asset prices? Most of the debate surrounding this question has focused on the ability of interest rate policy to prevent or control the appearance and growth of bubbles. But there is an alternative aspect of policy that was central to all of these episodes: the collapse of asset prices was accompanied by a fall into a liquidity trap and a substantial growth of central bank balance sheets. As the value of private assets evaporated, market participants turned to central banks for stores of value, which supplied them by expanding the monetary base (mostly in the form of reserves, but also through cash). Both in the US and in the euro area, for instance, the monetary base grew approximately fivefold in the aftermath of the financial crisis. This suggests that a key aspect of monetary policy in dealing with bubbles and their aftermath has been precisely to supply stores of value. Yet this role is completely absent in the New Keynesian paradigm that has dominated monetary economics over the last few decades, which emphasises the role of money as a unit of account and of nominal rigidities as drivers of monetary transmission (e.g. Gali 2009, Woodford 2011).
Without denying the usefulness of this paradigm, the events outlined above suggest that a shift in perspective may be called for. In a bubbly world, we can no longer disregard the role of money as a store of value, and the role of monetary policy as a supplier of stores of value. This raises a number of fundamental questions. When is money valuable as a store of value? How is this value connected to the rise and bursting of asset bubbles? Can the central bank always supply stores of value? If so, how much should it supply? In a recent paper (Asriyan et al. 2019), we develop a framework to address these questions.
To be sure, we are not the first to think of money as a store of value, as there is an old tradition in economics of adopting this perspective (e.g. Wallace 1981). We build on that tradition and its insights but combine it with the recent literature on asset bubbles and financial frictions (e.g. Martin and Ventura 2018). The framework that emerges provides a unified view on the connection between low interest rates, booms and busts in asset prices, and the role of money as a store of value. In particular, it enables us to study the similarities and differences between money and bubbles, their interaction, and the conduct of monetary policy in a bubbly world. In order to make the results as stark as possible, we completely neglect the role played by money as a unit of account and by nominal rigidities emphasized in the literature. Nonetheless, as we show, monetary policy retains a powerful role.
Bubbles and money: The view
The main elements of this view are easily explained. Consider an overlapping-generations world, in which some agents (entrepreneurs) want to borrow because they have productive investment opportunities, and other agents (savers) want to save because they do not have them. Normally, entrepreneurs would supply ‘backed’ assets, i.e. assets backed by the fruits of their investment, and savers would demand these assets as stores of value. But what if financial frictions restrict the supply of backed assets? Since savers still demand stores of value, these frictions depress the interest rate and open the door for ‘unbacked’ assets to be issued, i.e. assets that are backed only by expectations of their future value. Unbacked assets can be thought of as pyramid schemes, in which present contributions to the schemes (present purchases of the asset) give the right to receive future contributions (future purchases of the asset). As long as the expected return to these assets or pyramid schemes is sufficiently high, agents will be willing to hold them in equilibrium.
The dynamics of unbacked assets are driven by two forces, with differing effects on economic activity. First, their creation generates a wealth effect. New unbacked assets generate a rent for their creators because they are costless to produce and yet they have positive market value. For example, if an entrepreneur issues debt that is unbacked because the market expects it to be rolled over indefinitely, then she receives a pure rent. Second, the existence of unbacked assets generates an overhang effect. Old unbacked assets must be purchased and this diverts resources that could have been used for productive investment. In our example, the savers that actually finance the roll-over of the entrepreneur's debt must divert their funds from other uses. In short, the wealth effect of unbacked assets depends on their issuer, whereas the overhang effect depends on their buyer.
We consider two types of unbacked assets. The first is created by private entrepreneurs, and we refer to it as a bubble. The second is created by a public central bank, and we refer to it as money. Money can be valued as an asset if its expected rate of return (the inverse of the rate of inflation) equals that of the market (the real interest rate) – in other words, if the economy is in a liquidity trap. Both bubbles and money have wealth and overhang effects, but they differ in two crucial respects. First, bubbles are superior to money because their wealth effect accrues to entrepreneurs and, as a result, they foster investment. The wealth effect of money (i.e. seigniorage) accrues instead to the central bank, and its effect depends on how it is distributed among economic agents. Second, bubbles are inferior to money because they are driven by market psychology, so that their behaviour may be volatile and unpredictable, whereas money is under the control of the central bank. Ultimately, nothing guarantees that the size of bubbles supplied by the market will be optimal.
The role of monetary policy
We analyse the role of monetary policy in this bubbly world. To constrain this role as much as possible, we restrict the central bank's actions along three key dimensions. First, we assume that the central bank lacks fiscal backing, so that it cannot use tax proceeds to back the value of money. Second, we assume that the central bank is constrained in its use of seigniorage, so that it cannot choose how to distribute this revenue between entrepreneurs and savers. Finally, we assume that the central bank cannot affect market psychology, so that it must take the evolution of private bubbles as given. Despite these restrictions, we show that monetary policy has the ability to supply unbacked assets and – through this ability – plays a powerful role.
We find two main results.
- First, we show that the central bank can always intervene in the bubbly world, adjusting the money supply to provide unbacked assets over and above those supplied by private bubbles. Should it choose to do so, moreover, we show that the central bank can fully stabilize the economy's total supply of unbacked assets at a target of its choice! The choice of target is constrained by the market psychology that governs private bubbles, however. In particular, the central bank can only add to – but not subtract from – the unbacked assets supplied by private bubbles, and it is limited to implementing policies that guarantee non-explosive paths for these bubbles.
- Second, we show that the central bank should intervene in the bubbly world. We characterise an optimal monetary policy that raises the consumption of all generations along all histories by adjusting the supply of unbacked assets. To derive this result, we identify sequences of unbacked assets that are Pareto optimal, in the sense that – given any such sequence – it is impossible to raise the consumption of any one generation without reducing it for some other generation. The intuition here is the familiar one, i.e., unbacked assets are beneficial insofar as their overhang effect crowds out dynamically inefficient investments. By adjusting its supply of unbacked assets, monetary policy can ensure Pareto optimality.
The main takeaway of the theory is that central banks have a key role to play in the bubbly world: to supply assets. This resonates well with the conduct of monetary policy in the wake of the global financial crisis, when central banks expanded their balance sheets in response to the bursting of bubbles. But it is also quite different. In a standard balance sheet expansion, the central bank issues some assets and purchases others, leaving the net supply of assets available to the private sector unchanged. In a bubbly world, instead, the key aspect of welfare-enhancing interventions is that they expand the net supply of assets available to the private sector, thereby soaking up inefficient investment.
Our paper provides a macroeconomic framework of rational bubbles, credit, and monetary policy. A central feature of the framework is the presence of financial frictions, which limit the supply of backed assets and create space for unbacked assets to emerge. Our main finding is that, in this world, monetary policy plays a key role by expanding and stabilising the supply of unbacked assets at an optimal level.
Our analysis presents a first step towards understanding the optimal conduct of monetary policy in a bubbly world. However, much is still to be done. Our framework abstracts from nominal rigidities, which are prevalent in New Keynesian models. How would the introduction of such rigidities affect our results? Do they generate a trade-off between the optimal provision of assets by the central bank and the traditional price-stability objective of monetary policy? If so, under which conditions is one objective likely to dominate over the other? Our framework is a useful starting point for future research aiming at answering these crucial questions.
Authors’ note: The views expressed in this column are those of the authors and do not necessarily reflect the views of the ECB.
Asriyan, V, L Laeven and A Martin. (2019), “Collateral Booms and Information Depletion”, ECB Working Paper No. 2266.
Gali, J (2009), Monetary Policy, Inflation, and the Business Cycle: An Introduction to the New Keynesian Framework, Princeton University Press.
Martin, A and J Ventura (2018), “The Macroeconomics of Rational Bubbles: A User's Guide”, The Annual Review of Economics 10: 505-539.
Wallace, N (1981), “A Modigliani-Miller Theorem for Open-Market Operations”, The American Economic Review 71 (3): 267-274.
Woodford, M (2011), Interest and Prices: Foundations of a Theory of Monetary Policy, Princeton University Press.