MACROECONOMIC CONSEQUENCES OF DIGITAL TRANSFORMATION
DIGITAL TRANSFORMATION OF THE PAYMENT SYSTEM
Payment systems have evolved, and digitization has been an important contributor. Transfers are now more rapid as real-time gross settlement systems have been introduced in many jurisdictions for wholesale payments and fast payments for retail transactions are becoming more common. The decreased cost of cross-border transfers has been noted as one of the important potential benefits of using financial technology. This is the case for the cost of remittances. According to World Bank data quoted by the Economist (2019), in 2018 the average costs of sending $200 through banks and FinTech firms were 8% and 4%, respectively, with traditional money transfer firms lying in between. These differences lead to huge benefits for countries like the Philippines where remittances amount to close to 10% of GDP.
Financial technology applied to payments has also contributed significantly to financial inclusion, whereby formerly non-banked individuals and households have not only been able to carry out transactions via mobile phones, but also access simple financial services in the form of placing deposits and receiving loans. The systemic financial stability risks of the entrance of FinTech companies operating in the payment system are modest if their operations are limited to facilitating transactions, and as long as they do not constitute a substantial portion of the overall payment system. If payment innovators use their position to enter credit extension activities this benign situation may change, particularly if they remain unregulated as lenders. If they encroach on the credit extension business of traditional financial intermediaries, the franchise value of the latter may be threatened, inducing them to pursue riskier lending to preserve their market share. As with the entrance of Big-Tech companies in the credit extension business, regulators will need to monitor payment innovators and regulate them on an activity basis rather than on an institutional basis.
While FinTech payment companies do not constitute a threat to the business model of banks, another innovation made possible by advances in technology might: cryptocurrencies. The threat is not from the current collection of privately issued cryptocurrencies such as Bitcoin and the like. These should not be considered money in the conventional sense of being units of account, means of payment, and stores of value. They are speculative assets, and as such they do not encroach significantly on intermediation activities of banks and other financial institutions. The Libra project floated by Facebook could have had an impact on the structure of banking, not principally through its role as a payment vehicle but rather as a means for Facebook to enter the financial services industry providing deposit, lending, and asset management facilities. Unlike Bitcoin and other similar cryptocurrencies, Libra promised to maintain a stable value relative to a government fiat currency or a basket of such currencies. This constitutes its fatal flaw, because we have learned from countless efforts to fix currency values that success in doing so requires holding inventories of the fiat currency that are at least as large as the outstanding stock of the Libra. Combined with the unreceptive attitude of central banks and regulators towards the project, the threat from Libra to banks is not likely to be significant. A threat to bank intermediation could, however, come from crypto-, or digital, currencies issued by central banks, so-called “Central Bank Digital Currencies,” or CBCDs. These are digital means of payments guaranteed in value by the central bank in terms of the domestic fiat currency. The central bank can live up to this promise because it is the issuer of the domestic currency. The CBDC can take one of two forms: a token-based form that is issued by the central bank but managed by the banking system, or an account-based form whereby individuals would have accounts with the central bank denominated in the CBDC. It is the second form that could lead to substantial challenges for the private sector intermediation system. Since accounts with the central bank would be less risky than accounts with commercial banks, there is a risk that such a system would lead to disintermediation of the banking system. For this reason, it is likely that the implementation of any CBDC would be a token-based model.
MACROECONOMIC CONSEQUENCES OF DIGITAL TRANSFORMATION
Digital transformation is also taking place in the nonfinancial sectors of the economy. Industrial robots are increasingly used in manufacturing, online commerce has been growing rapidly, globalization of production chains has upended traditional production processes, etc. These changes have important implications for how we should think about prospects for economic growth and for the sources and consequences of macroeconomic fluctuations and what the appropriate policy responses should be.
Will Digital Transformation Bring about a New Industrial Revolution?
The conventional way to think about economic growth is in terms of a relationship between the economy’s output per capita and inputs such as physical and human capital per capita on the one hand and a residual that captures the state of the available technology, broadly defined, on the other. Economic growth in this framework can come about through investment in physical and human capital in excess of population growth and technological progress. In a balanced growth environment where investments are just enough to equip the growing labor force with the existing amount of capital and skills, growth will come about exclusively through technological progress, or “total factor productivity” as it is called in the technical literature. The issue of whether digital transformation will bring about a burst of economic growth then hinges on its effect on this total factor productivity.
Nobel Laureate Robert Solow famously said about the effect of computers on economic growth that “you can see the computer age everywhere but in the productivity statistics,” suggesting that early digitalization did not lead to measurable increases in growth (Solow 1987). Could it be that the effects of more recent developments in robotics, autonomous vehicles, and internet availability will also fail to show up in productivity statistics? One of the proponents of this view is Robert Gordon, who points to the decline in total factor productivity (TFP) growth in the United States since 1970 to less than 1% per year, following the historically high growth rate of close to 2% during the 50 years from 1920 to 1970. According to Gordon, the strong TFP growth in that period was the consequence of the application of electricity in manufacturing, transport, and communication, the internal combustion engine, and devices such as the telephone and the radio, and to the increasing availability of running water and sewage that improved health and longevity. Gordon’s views have not gone unchallenged. According to some authors, digital transformation of the entire economy, not just the financial sector – industrial robots, self-driving cars and trucks, “the internet of things” – driven by AI has the potential to increase productivity far beyond what we have seen in the recent past. According to this view, a number of features of the current wave of digital transformation are said to be different from past ICT developments and will therefore have a greater effect on productivity and well-being. First, much of the current digital transformation is about the production of ideas rather than the production of goods. Furthermore, ideas are public goods, and because a large proportion of them are shared on the internet, their reach is global and available for anyone to build on and improve. Networks of innovators can be formed, and the shared knowledge and progress will be much more rapid and widespread, spurring additional ideas. Second, many of the internet-based services that have great value (e.g., online shopping, online translation) are available free of charge and are therefore not measured in GDP and hence in productivity statistics even though they are of great value. In contrast to these optimistic scenarios, the easy scalability of digital transformations may contain seeds whose effects are less benign. Combined with large setup costs and very low marginal costs of expansion, it is possible that concentration and market dominance will emerge. That in turn could lead to increased inequality and reduced motives for innovation.
Stabilization policy is by definition concerned with short- to medium-term fluctuations in economic activity around its longer-run potential path. If the longer-run path is uncertain, so will be any measure of the deviation therefrom. If monetary and fiscal policies are calibrated to erroneous measures of economic activity gaps, the resulting policy stance will be inappropriate. A related issue concerns modeling, and particularly standard modeling of inflation. The conventional approach is to relate inflation pressures to some measure of slack in the economy, usually measured by an output gap measure. Could it be that difficulties encountered in explaining inflation are due to difficulties in measuring the relevant output gap? More research on these issues is warranted.
Global Value Chains and Monetary Policy
This is being written as the Coronavirus pandemic is ravaging the globe. The economic fallout of the pandemic and the measures taken in attempts to contain it have brought home how immensely interwoven a country’s different economic regions and sectors are. A manufacturing firm in one region has to curtail its operations because it cannot obtain the required parts that are produced by another firm in another region. As a consequence, our firm is unable to ship its products that are crucial for the assembly of the final product, which is carried out in yet another region by yet another firm. Replace regions in this example with countries in the global economy and we have a description of global value chains (GVCs). The importance of these GVCs has increased over time as digitalization has facilitated fragmentation and outsourcing of production to take advantage of different comparative advantages across economies and economies of scale in production. The implications of production chains, and global value chains in particular, is that they tie regions and economies more closely together. A dislocation in supply in one part of the chain, a supply shock, will propagate throughout the system causing a decrease in output in the system as a whole, be it the entire domestic economy in the case of domestic supply chains or the world as a whole in the case of GVCs. Likewise, an increase in demand for the final output will propagate throughout the production system as the demand for intermediate inputs will increase. Price developments will also become more global, as the price of the final output will depend not only on the cost of production locally but on the additions to cost throughout the value chain. Inflation rates will become more closely linked across economies.
What are the implications for macroeconomic stability and macroeconomic stabilization policies?
The most obvious is that greater connectedness means that policy spillovers will be stronger, making some form of policy consultation between country authorities more desirable. To some extent this is already taking place in fora such as the G20, the BIS, and the IMF. While formal policy coordination may not be achievable because authorities in every country are accountable to their own constituents, some tacit agreement to proscribe polices that directly harm others, so-called “beggar-thy-neighbor policies,” would be desirable. At the national level, the increased importance of GVCs appears to have changed inflation dynamics. A study by Auer, Borio, and Filardo at the BIS indicates that as GVCs have increased in importance, the role of the domestic output gap in domestic inflation has decreased, and the role of a measure of a global output gap increased (Auer, Borio, and Filardo 2017). There are also studies suggesting that the Phillips curve has become flatter, i.e., that the coefficient on domestic economic slack (or overheating) in an estimated inflation equation has become smaller (Carney 2017). If his last finding is indeed a feature of the new economic structure central banks are facing, it begs the question as to how they can hope to implement inflation targeting strategies successfully.
REFLECTIONS ON THE POSSIBLE IMPLICATIONS OF THE CORONAVIRUS PANDEMIC
As already noted, this is being written in the midst of the 2020 Coronavirus pandemic that has created unimaginable human suffering and great economic upheaval. As it unfolds, it is hard to imagine that the world will return to what it was only half a year ago.
A salient feature of the digital transformation of finance is that virtual AI-assisted financial intermediation is challenging financial intermediation and payment services that are based on personal contacts. The social-distancing behavior that has been mandated or highly recommended during the pandemic increases the competitive advantage of the virtual business model. Entities that have a very broad access to potential customers, either through their social media presence or their internet-based commerce engagement, will be particularly strongly positioned to expand in this environment. These are the Big-Tech firms.
Regulatory authorities must be vigilant and make sure that the financial services activities of these firms are appropriately regulated. The discussion on the macroeconomic stability effects of digital transformation focused primarily on the consequences of increased interconnectedness brought about by production chains, both regional and global. Some of the economic havoc brought about by the Coronavirus pandemic is the result of this interconnectedness. The business case for some fragmentation of production processes will certainly remain, but one cannot rule out that there will be some retrenchment. This may give back some of the effectiveness of domestic economic policies lost to global influences, but it would also reduce some of the benefits from trade, which would be particularly painful for small trade-dependent developing and emerging economies.
REFERENCES
Hans Genberg, Digital Transformation, Working Paper, May, 2020
Auer, Raphael, Claudio Borio, and Andrew Filardo (2017). “The globalisation of inflation: the growing importance of global value chains”, BIS Working Paper No. 602.
Bank for International Settlements (2019). “Big tech in finance: opportunities and risks”, in BIS Annual Economic Report 2019, pp. 55–79.
Brynjolfsson E. and A. McAfee (2014). The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies, New York: W.W. Norton & Company.
Cambridge Centre for Alternative Finance (2020), Transforming Paradigms: A Global AI in Financial Services Survey. https://www.jbs.cam.ac.uk/fileadmin/ user_upload/research/centres/alternative-finance/downloads/2020-ccaf-ai-infinancial-services-survey.pdf.
CCAF, ADBI, FinTechSpace (2019). ASEAN FinTech Ecosystem Benchmarking Study. Cambridge, UK.
Carney, Mark (2017). 2017 Michel Camdessus Central Banking Lecture, International Monetary Fund, https://www.imf.org/en/News/Events/2017-michel-camdessuscentral-banking-lecture.
Danielsson, Jon, Rogert Macrae, and Andreas Uthemann (2017). “Artificial Intelligence, financial risk management and systematic risk”, SRC Special Paper No 13, November. Accessed at https://www.systemicrisk.ac.uk/sites/default/files/ downloads/publications/SP13.pdf.
Diaz-Alejandro, Carlos F (1985) “Good-bye financial repression, hello financial crash”, Journal of Development Economics 19, 1–24.
Economist, The (2019). “Fintech takes aim at the steep cost of international money transfers”, 11 April edition.
Financial Stability Board (2019). “Big-Tech in Finance: Market developments and potential financial stability implications”, 9 December. Accessed at https://www.fsb.org/wp-content/uploads/P091219-1.pdf.
Gordon, Robert J. (2015). “The Sources of Slowing Growth in Productivity Growth and Potential Output”, Presentation at the Northwestern University and NBER – Philadelphia Fed Policy Forum, 4 December 2015. https://www.philadelphiafed.org/-/media/research-and-data/events/2015/ fed-policy-forum/papers
World Bank (2020). Trading for Development in a World of Global Value Chains. Washington, DC, International Bank for Reconstruction and Development/The World Bank.