Technology risks: 17x new research on SDGs, nuclear, blockchain and AI risks, innovation, climate, carbon offsets, ESG ratings, treasuries, backtests and trading, big data, forensic finance, private equity and other alternatives by Patrick Behr, Richard Ennis, Christian von Hirschhausen, Thierry Roncalli, Bernhard Schwetzler and many more (# shows SSRN downloads on August 17th, 2023):
Social and ecological research (Technology risks)
SDG or green? Take a Deep Breath! The Role of Meeting SDGs With Regard to Air Pollution in EU and ASEAN Countries by Huynh Truong Thi Ngoc, Florian Horky, and Chi Le Quoc as of July 10th, 2023 (#26): “First, the results show that in ASEAN countries, Goal 10 (Reduced Inequalities) has a negative correlation with most other SDGs while in the EU it shows a broadly positive correlation. … air pollution, particularly SO2 and CO emissions, is positively connected to most SDGs in ASEAN while the trend in the EU is not clear. This could be due to the rapid economic development in ASEAN nations as well …” (p. 19).
Nuclear risks: The Potential of Nuclear Power in the Context of Climate Change Mitigation -A Techno-Economic Reactor Technology Assessment by Fanny Böse, Alexander Wimmers, Björn Steigerwald, and Christian von Hirschhausen as of July 27th, 2023 (#17): “… we synthesize techno-economic aspects of potential new nuclear power plants differentiating between three different reactor technology types: light-water cooled reactors with high capacities (in the range or above 1,000 MWel), so-called SMRs (“small modular reactor”), i.e., light-water cooled reactors of lower power rating (< 300 MWel) (pursued, e.g., in the US, Canada, and the UK), and non-light water cooled reactors (“so-called new reactor” (SNR) concepts), focusing on sodium-cooled fast neutron reactors as well as high-temperature reactors. … Actual development .. shows an industry in decline and, if commercially available, lacking economic competitiveness in low-carbon energy markets for all reactor types. Literature shows that other reactor technologies are in the coming decades unlikely to be available on a scale that could impact climate change mitigation efforts. The techno-economic feasibility of nuclear power should thus be assessed more critically in future energy system scenarios“ (abstract).
Blockchain risks: On the Security of Optimistic Blockchain Mechanisms by Jiasun Li as of August 15th, 2023 (#68): “Many new blockchain applications … adopt an “optimistic” design, that is, the system proceeds as if all participants are well-behaving … We point out that such protocols cannot be secure if all participants are rational” (abstract). “Given that alternative solutions are still technically immature, … the community either has to deviate from its pursuit of decentralization and accept a system that relies on trusted entities, or accept that fact their systems cannot be 100% secure” (p. 33).
AI chains: Determining Our Future: How Artificial Intelligence Creates a Deterministic World by Yuval Goldfus and Niklas Eder as of Aug. 9th, 2023 (#22): “… we demonstrate that AI relies on a deterministic worldview, which contradicts our most fundamental cultural narratives. AI-based decision making systems turn predictions into self-fulfilling prophecies; not simply revealing the patterns underlying our world, but creating and enforcing them, to the detriment of the underprivileged, the exceptional, the unlikely. The widespread utilisation of AI dramatically aggravates the tension between the constraints of environment, society, and past behavior, and individuals’ ability to alter the course of their lives, and to be masters of their own fate. Exposing hidden costs of the economic exploitation of AI, the article facilitates a philosophical discussion on responsible uses. It provides foundations of an ethical principle which allows us to shape the employment of AI in a way which aligns with our narratives and values” (abstract). My comment: My opinion regarding AI for sustainable investments see How can sustainable investors benefit from artificial intelligence? – GITEX Impact – Leading ESG Event 2023
Musical therapy? The Value of Openness by Joshua Della Vedova, Stephan Siegel, and Mitch Warachka as of July 5th (#48): “We construct a novel proxy for openness using MSA-level data (Sö: US Metropolitan Statistical Areas) from radio station playlists. This proxy is based on the adoption of new music and varies significantly across MSAs. Empirically, we find a robust positive association between openness and proxies of value creation such as the number of new ventures funded by venture capital, the number of successful exits by new ventures, the proportion of growth firms, and Tobin’s q. … An instrumental variables procedure confirms that openness is highly persistent with variation across MSAs being evident more than a century before the start of our sample period. … our results are especially strong for young firms that are more likely to depend on new products“ (p. 26/27).
ESG and impact investing research
Climate stress: From Climate Stress Testing to Climate Value-at-Risk: A Stochastic Approach by Baptiste Desnos, Théo Le Guenedal, Philippe Morais, and Thierry Roncalli from Amundi as of July 5th, 2023 (697): „This paper proposes a comprehensive climate stress testing approach to measure the impact of transition risk on investment portfolios. … our framework considers a bottom-up approach and is mainly relevant for the asset management industry. … we model the distribution function of the carbon tax, provide an explicit specification of indirect carbon emissions in the supply chain, introduce pass-through mechanisms of carbon prices, and compute the probability distribution of potential (economic and financial) impacts in a Monte Carlo setting. Rather than using a single or limited set of scenarios, we use a probabilistic approach to generate thousands of simulated pathways” (abstract).
Disaster flows: Flight to climatic safety: local natural disasters and global portfolio flows by Fabrizio Ferriani, Andrea Gazzani, and Filippo Natoli from the Bank of Italy as of July 5th, 2023 (#35): “… we find that local natural disasters have significant effects on global portfolio flows. First, when disasters strike, international investors reduce their net flows to equity mutual funds exposed to affected countries. This only happens when disasters occur in the emerging economies that are more exposed to climate risk. Second, natural disasters lead investors to reduce their portfolio flows into unaffected, high-climate-risk countries in the same region as well. Third, disasters in high-climate-risk emerging economies spur investment flows into advanced countries that are relatively safer from a climate risk standpoint“ (abstract).
Carbon offsets: Portfolio Allocation and Optimization with Carbon Offsets: Is it Worth the While? by Patrick Behr, Carsten Mueller, and Papa Orge as of Aug. 10th, 2023: “We explore whether the integration of carbon offsets into investment portfolios improves performance. … our results show that investment strategies that include such offsets broadly achieve higher Sharpe Ratios than the diversified benchmark, with the long-short strategy performing best”.
Useless ratings? ESG Ratings Management by Jess Cornaggia and Kimberly Cornaggia as of July 27th, 2023 (#92): “We use data from an ESG rater that incorporates feedback from firms during the rating process and produces ratings at a monthly frequency. We … find that when the rater changes the weight it applies to certain criteria in the creation of its ESG ratings, firms respond by adjusting their reported ESG behavior in the same month. … we do not observe real changes in the likelihood that firms are embroiled in ESG controversies, or that they reduce their release of toxic chemicals because of these adjustments. Rather, it appears firms “manage” their ESG ratings for the benefit of ESG-conscious investors and customers” (p. 26/27). My comment: I do not use market leading MSCI or ISS or Sustainalytics ratings and also because of my custom rating profile (Best-in-Universe with specific approach to treat missing data) the risk of such ratings management should be low, see Noch eine Fondsboutique? – Responsible Investment Research Blog (prof-soehnholz.com)
AI and other investment research (Technology risks)
ETFs effect Treasuries: ETF Dividend Cycles by Pekka Honkanen, Yapei Zhang, and Tong Zhou as of Aug. 10th, 2023 (#340): “… in the “ETF dividend cycle,” ETFs accumulate incoming corporate dividends in MMFs (Sö: Money Market Funds) gradually but withdraw them abruptly in large amounts when they themselves have to pay dividends to investors. … This … leads to large, sudden outflows from MMFs, forcing these funds to liquidate some of their underlying assets. We find that these liquidations are concentrated in short-term Treasury bonds. … in the aggregate time series, an ETF dividend distribution event of average size leads to increases in short-term Treasury yields by approximately 0.38-0.58 basis points. … The total value fluctuation in the Treasury market could be considerable, as ETFs distribute dividends on 205 trading days in 2019, for example” (p. 9/10).
Backtest-problems: Market Returns Are Estimated with Error. How Much Error? by Edward F. McQuarrie as of July 24th, 2023 (#30): “For periods beginning 1926, it is conventional to suppose that historical market returns are known with reasonable accuracy. This paper challenges that comfortable certainty. Multiple indexes of market return are examined to show that return estimates do not closely agree across indexes and are unstable within index over time. The paper concludes that two-decimal precision—to the whole percentage point, with an error band of plus or minus one percentage point—would better reflect the accuracy of historical estimates of annual market return” (abstract).
Easy profits: Intraday Stock Predictability Everywhere by Fred Liu, and Lars Stentoft as of July 5th, 2023 (#1167): “First, we demonstrate that the market and sector portfolios are highly predictable. … we show that portfolio profitability mostly remains high after accounting for transaction costs, and is largely orthogonal to common risk factors. … we further exploit machine learning forecasts of individual stocks by constructing machine learning intraday portfolios, and demonstrate that a long-short portfolio achieves a Sharpe ratio of up to 4 after transaction costs. … demonstrate that less liquid firms are more predictable and firms which are more actively traded and volatile tend to be more profitable … intraday predictability and profitability generally decrease as the time horizon increases” (p. 28/29). My comment: If this is so easy, why do Quant funds typically disappoint? The information is important for stock trading, though (for my trading approach see Artikel 9 Fonds: Sind 50% Turnover ok? – Responsible Investment Research Blog (prof-soehnholz.com)).
Satellite vs. people: Displaced by Big Data: Evidence from Active Fund Managers by Maxime Bonelli and Thierry Foucault as of Aug. 2nd, 2023 (#325): “We test whether the availability of satellite imagery data tracking retailer firms’ parking lots affects the stock picking abilities of active mutual fund managers in stocks covered by this data. … we find that active mutual funds’ stock picking ability declines in covered stocks after the introduction of satellite imagery data for these stocks. This decline is particularly pronounced for funds that heavily rely on traditional sources of expertise, indicating that these managers are at a higher risk of being displaced by new data sources“ (p. 29/30).
AI bubble? Artificial Intelligence in Finance: Valuations and Opportunities by Yosef Bonaparte as of August 15th, 2023 (#65): “First, we display the current and projected AI revenue by sector, technology type, and geography. Second, present valuation model to AI stocks and ETFs that accounts for AI sentiment as well as fundamental analyses. Our findings demonstrate that the AI revenue will pass $2.7 trillion in the next 10 years, where the service AI technology stack will contain 75% of the market share (as of 2023 it is 50% of the market share). As for AI stock valuation, we present two main models to adopt when we value stocks“ (abstract).
Bad finance: What is Forensic Finance? by John M. Griffin and Samuel Kruger as of Aug. 10th, 2023 (#467): “We survey a growing field studying aspects of finance that are potentially illegal, illicit, or immoral. Some of the literature is investigative in nature to uncover malfeasance that is recent and possibly ongoing. … The work spans newer areas such as cryptocurrencies, financial advisor and broker misconduct, and greenwashing; and newer research in established fields that are still developing, such as insider trading, structured finance, market manipulation, political connections, public finance, and corporate fraud. We highlight investigative forensic finance, common economic questions, common empirical methods, industry and political opposition, censoring, and the importance of avoiding publication biases“ (abstract).
Specialist PE: Specialization in Private Equity and Corporate Financial Distress by Benjamin Hammer, Robert Loos, Lukas Andreas Oswald, and Bernhard Schwetzler as of Aug. 7th, 2023 (#384): “We investigate the impact of industry specialization of private equity firms on financial distress risk of portfolio companies … Difference-in-differences estimates suggest an increase in distress risk through private equity backing. The effect is stronger for specialist-backed firms than for generalist-backed firms relative to a carefully matched control group. However, specialist-backed firms can afford the increase in distress risk because they are less risky than generalist-backed firms before the buyout. Overall, our findings are consistent with the idea that greater idiosyncratic risk in specialized PE portfolios induces more risk-averse target selection” (abstract).
Costly diversification: Have Alternative Investments Helped or Hurt? by Richard M. Ennis as of August 3rd, 2023 (#135): “This paper shows that since the GFC (Sö: Global Financial Crisis in 2007/2008), US public-sector pension funds’ exposure to alternative investments is strongly associated with a reduction in alpha of approximately 1.2 percentage points per year relative to passive investment. While exposure to private equity has arguably neither helped nor hurt, both real estate and hedge fund exposures have detracted significantly from performance. Institutional investors should consider whether continuing to invest in alts warrants the time, expense and reduced liquidity associated with them” (p. 11).
Advert for German investors:
Sponsor my research by investing in and/or recommending my global small/midcap mutual fund (SFDR Art. 9). The fund focuses on social SDGs and uses separate E, S and G best-in-universe minimum ratings and broad shareholder engagement with currently 28 of 30 engaged companies: FutureVest Equity Sustainable Development Goals R – DE000A2P37T6 – A2P37T; also see Active or impact investing? – (prof-soehnholz.com)