LeituraPesquisaRadarFramework de investimento
Entrar / Cadastrar
TudoTecnologiaProdutosInvestimentoBolsaOutros

Leitura

2026-11-167 posts

Atualizar

Federal Reserve Bank of New York Research: Post-Pandemic Workers Systematically Underestimate Job-Finding Probability, with Largest Bias Among Youth and Low-Education Groups

Tópico · 纽约联储劳动力误判研究Material
2026-11-16 20:00 GMT+8

In research published on November 16, 2026, a team from the Federal Reserve Bank of New York—including Richard Audoly, Ivy Chen, Miles Guerin, and Rachel Schuh—systematically compared subjective estimates from the Survey of Consumer Expectations (SCE) regarding "the likelihood of finding a new job within three months if you lost your current job" with actual re-employment rates tracked in the Current Population Survey (CPS). The results showed high alignment between expectations and reality from 2014 to 2019: the average expected three-month re-employment rate was 55.5%, compared to an actual rate of 57.2%. However, a persistent gap emerged after 2020: from 2022 to the present, the actual re-employment rate has averaged 58.8%, while worker expectations have dropped to 53.0%.

This pessimism is not evenly distributed. Workers aged 30 and older and those with a college education saw little change before and after the pandemic, consistently remaining slightly below actual rates (by about 4 to 6 percentage points). In contrast, workers under 30 experienced a significant reversal: pre-pandemic they were optimistic, with expectations exceeding actuals by 6.3 percentage points; post-pandemic they turned pessimistic, underestimating their re-employment probability by 3.0 percentage points. Workers with a high school diploma or less exhibited the largest negative bias, with expectations 7.5 percentage points lower than actuals. The authors note that while rising youth unemployment is a frequently discussed topic, overly negative expectations themselves may exacerbate the downbeat employment outlook for new labor market entrants.

On the side of job-loss risk, the study found that workers generally expect to lose their jobs at higher rates than actually occur. When converting the SCE's twelve-month expected probabilities to a three-month horizon, the actual job-loss probability is approximately half of what is expected. More notably, this gap widens with tenure: employees who have been with their current employer for more than two years expect a job-loss probability more than twice the actual rate. The authors argue this reflects workers underestimating the job security provided by tenure—the actual decline in job-loss risk as tenure increases is far greater than workers perceive.

For investors, this research highlights a quantifiable measurement bias: if relying on survey data to gauge labor market tightness, one must recognize that worker expectations are systematically pessimistic, particularly among younger and less-educated groups. This implies that the actual labor market may be more resilient than sentiment indicators suggest, and downward pressure on wages may be less severe than surface data indicates. However, the authors acknowledge that their CPS-based measure of unemployment is a lower bound—it may miss individuals who exit the labor force or quit voluntarily before being laid off—meaning part of the gap between expectations and reality may reflect differences in measurement methodology rather than pure cognitive bias.

Fontes:Liberty Street Economics

Pesquisa

The Emission Reduction Impact of Net-Zero Commitments: Three Years of Data from 1,200 Major Companies Shows No Significant Effect

Tópico · 净零承诺减排效果Material
2026-11-16 15:01 GMT+8

At the end of 2019, only 70 out of roughly 2,000 major global companies had committed to long-term net-zero targets. By the end of 2024, this number had risen to more than 1,200 companies, accounting for two-thirds of the combined market capitalization of the sample. The proliferation of net-zero commitments has been astonishing, but have these commitments themselves led to measurable emission reductions? A study published in 2026 by Simon Dietz and Nikolaus Hastreiter provides a statistical test of this question.

The researchers compared changes in absolute emissions and emission intensity before and after companies adopted net-zero targets. In terms of direction, the results are encouraging: three years after adopting a target, average absolute emissions decreased by about 15%, and average emission intensity dropped by nearly 10%. The trend is indeed moving in the right direction. However, the magnitude of this change is insufficient to reach statistical significance. In other words, we cannot rule out the possibility that these declines were driven by random fluctuations or other concurrent factors.

For investors, this means that the environmental impact of current corporate net-zero commitments lacks rigorous statistical evidence. The spread of commitments alone does not equate to actual emission reductions. Of course, this conclusion has a clear temporal boundary: net-zero targets are typically set for 2040 to 2050, and the study only observed data for three years post-adoption. As author Joachim Klement pointed out, as emission reductions increase over time, statistical significance may change; but at least for now, we cannot conclude that net-zero commitments have significantly reduced greenhouse gas emissions.

Fontes:klementoninvesting.substack.com

Pesquisa

Greenpeace Activist Benjamin Stephan: Why AI Expansion Accountability Points to Nvidia

Tópico · 绿色和平问责英伟达Material
2026-11-16 18:00 GMT+8

The expansion of AI data centers is encountering increasingly strong social resistance. Data Center Knowledge reports that residents, local protest groups, and large organizations are all expressing dissatisfaction with data centers. Greenpeace has chosen a path different from typical community opposition: holding chip supplier Nvidia directly accountable. In an interview with Data Center Knowledge, Benjamin Stephan, Greenpeace International’s Big Tech campaigner, fully elaborated on the logic behind this strategic choice.

Stephan’s argument does not start with the broad criticism that 'data centers consume too much electricity,' but rather focuses on Nvidia’s unique position within the economic chain of AI infrastructure. He points out that Nvidia now does more than just sell chips; it organizes financing models that make it easier for major tech companies to secure funding to build new data centers and purchase its chips, further fueling unchecked data center expansion. In Stephan’s view, this means Nvidia has transformed from a mere supplier into a driver and beneficiary of expansion, making it an accountable node.

Another link in the accountability chain comes from the supply chain. Stephan notes that Nvidia’s supply chain largely runs on fossil fuels, exacerbating the climate crisis while creating significant hardship for communities in Taiwan, where Nvidia relies on TSMC to manufacture its chips. He argues that given Nvidia’s profit scale, switching its supply chain to renewable energy would cost only a fraction of its earnings.

The third link involves military use. Stephan cites reports that Nvidia chips have been found in Russian autonomous drones deployed in Ukraine, deeming this unacceptable. He demands that Nvidia implement independent audits across all global distributors, subsidiaries, and third-party vendors to ensure that no Nvidia technology is used to violate human rights or diverted into military equipment and autonomous weapons.

These three threads converge on a specific event: the Taipei promise. Stephan recalls that at a previous GTC conference in Taipei, Greenpeace East Asia activists presented Huang with a five-layer cake topped with wind turbines, borrowing his metaphor that 'AI is a five-layered cake' to point out that AI and Nvidia were missing the layer of renewable energy. When asked if he could invest in renewable energy together with Taiwanese suppliers, Huang gave a positive response. Stephan said, 'We were in Berlin to hold him accountable because nothing had happened.'

The specifics of the Berlin protest also corroborate this accountability logic. Stephan, who was present as a spokesperson, waited outside by Huang’s car after the CEO left the venue, informing security that Huang had said on stage they would 'talk later.' However, the vehicle drove off empty, and the conversation did not take place. Stephan emphasized that this was not an action organized by his group; as a Berlin-based activist, he was simply seeking a direct dialogue in his capacity as a spokesperson.

Fontes:DataCenterKnowledge

Pesquisa

Econbrowser Blogger Menzie Chinn: Four Recession-Dating Methods Yield Different Answers for Germany in 2025–26

Material
2026-11-16 07:50 GMT+8

Economic data for Germany in 2025–26 present a seemingly contradictory set of signals: employment numbers and hours worked are declining, yet real GDP is growing strongly. In a post published on November 15, 2026, Econbrowser blogger Menzie Chinn compared four different methods for dating recessions in Germany, finding that they yielded different answers for the same period.

The first method is from ECRI (The Conference Board's Economic Cycle Research Institute). The second is from the German Council of Economic Experts (GCEE), which follows the methodological framework of the U.S. National Bureau of Economic Research (NBER) and the Centre for Economic Policy Research (CEPR), dividing economic cycles into periods of expansion and contraction. The third is Laurent Ferrara’s 2019 method, which relies on GDP and industrial production data, using the Bry and Boschan algorithm to identify peaks. The fourth is Chinn’s own calculation based on pure employment data, primarily drawing from national accounts with minor reference to household surveys.

Chinn’s core argument is that it is difficult to argue Germany is in a recession given the backdrop of strong GDP growth. GDP-based methods (the GCEE and Ferrara) did not classify 2025–26 as a recession, whereas the pure-employment measure indicated contraction. The Ifo Institute’s nowcast released on November 6 showed that Germany’s Q4 GDP grew by 0.62% quarter-on-quarter, further supporting a judgment of expansion rather than contraction.

However, the source text only displays the differences in classification results among the various methods; it does not explain the economic reasons for the coexistence of falling employment and rising GDP, nor does it provide a comparison of the historical accuracy rates of these methods. Chinn’s conclusion relies on the premise that GDP growth will continue—if growth slows while employment continues to deteriorate, the classification conclusions could reverse. Additionally, Ferrara’s chronology predates the pandemic, and Chinn assumed in his graph that its 2020 dates matched those of ECRI, which is itself an unverified processing assumption.

Fontes:econbrowser.com

Pesquisa

Cross-Validation of Five Probability-Sampling Polls: American Negative Views on AI Are Robust and Cross-Partisan

Tópico · 美国公众AI负面看法Material
2026-11-16 08:00 GMT+8

In 2026, five high-quality polls using probability sampling methods—Pew Research Center, Gallup, Quinnipiac University Poll, the Annenberg Public Policy Center, and Ipsos—asked U.S. adults about their views on artificial intelligence (AI) using different wordings, option sets, and collection modes. The direction of results was completely consistent: every survey found that more Americans held negative views of AI than positive ones.

This conclusion itself is not surprising, but its robustness is noteworthy. Poll wording does affect specific numbers: when a middle option is provided, a large share of respondents select it—37% chose the middle option in Pew’s 2026 survey—while the positive option’s share is compressed to just 9%. Consequently, absolute proportions across different surveys cannot be directly compared. However, all five surveys point unerringly in the same direction toward the same judgment: public concern about AI in the United States outweighs anticipation or excitement.

Pew’s data particularly illustrates the shift in trends. In the June 2026 survey, 52% of respondents said they were “more concerned than excited” about AI, while only 9% said they were “more excited than concerned,” a ratio exceeding 5:1. When Pew first asked this same question in November 2021, the ratio was approximately 2:1. Over these four-plus years, the gap between concern and enthusiasm has more than doubled.

An Ipsos survey conducted in October 2026 further showed that when people hold strong feelings, the majority are negative: 17% considered AI an “extremely bad” thing for society, compared to just 2% who considered it “extremely good.” A Quinnipiac University Poll telephone survey from September 2026 touched upon the most extreme concerns: 73% of Americans believe AI poses a risk to human survival, with 38% expressing that they are very concerned.

This concern crosses partisan lines. Pew data shows that 56% of Democrats and 49% of Republicans are more concerned than excited about AI; in both parties, only about one in ten people are more excited than concerned. Anxiety regarding AI is not the exclusive emotion of any single political faction.

However, there is one area where results diverge from the above pattern: when asked about AI’s impact on personal life rather than society as a whole, the Ipsos survey found that 50% viewed it as a good thing and 49% as a bad thing, nearly splitting evenly. Yet even within this more personalized framework, strong feelings still lean negative—15% believed AI would be extremely harmful to them personally, versus only 3% who believed it would be extremely beneficial. In a Quinnipiac University Poll conducted in March 2026, 55% expected AI to cause more harm than good in their daily lives, while 34% expected more good than harm.

The core implication of this data set is that American public attitudes toward AI are not an illusion manufactured by questionnaire wording, but a stable pattern spanning methodology, time, and partisanship. If the AI industry hopes to win public trust, it faces not merely a public opinion issue that can be repackaged, but a widespread concern that has already taken root.

Fontes:Our World in Data

Pesquisa

InstructGPT Paper Summary: How Human Feedback Enables Language Models to Follow Instructions

Verificado 2026-11-16 16:14 GMT+8

Leitura de longo prazo · 《Training language models to follow instructions with human feedback》(2022)

This paper, published by the OpenAI team in 2022 and titled "Training language models to follow instructions with human feedback" (训练语言模型以遵循人类反馈的指令), raises a core question: why are larger language models not necessarily better at understanding user intent? The authors note that traditional large language models (such as GPT-3), despite their massive parameter counts, frequently generate unrealistic, toxic, or unhelpful content because their objective function is to predict the next word rather than "safely and helpfully follow user instructions." This misalignment between the model's objective and user intent is referred to as "misaligned."

To address this issue, the paper proposes a method called RLHF (Reinforcement Learning from Human Feedback). This method consists of three steps: first, supervised fine-tuning (SFT) of GPT-3 using prompts and demonstration data written by annotators to obtain a baseline model; second, collecting human preference ranking data for model outputs to train a reward model (RM) that predicts which output humans prefer; third, using this reward model as a signal to further fine-tune the supervised model via the PPO algorithm to maximize human preference scores. The resulting model is termed InstructGPT.

Experimental results show that in human evaluations, the InstructGPT model, with only 1.3B parameters, generated outputs more preferred than those of the 175B-parameter GPT-3, despite having 100 times fewer parameters. Additionally, InstructGPT showed significant improvements in truthfulness; for example, in the TruthfulQA benchmark, it generated true and informative answers approximately twice as often as GPT-3. However, the paper also notes limitations: InstructGPT still makes simple errors, such as fabricating facts or failing to detect instructions containing false premises; and there is performance regression on public NLP datasets, known as the "alignment tax," although this can be mitigated through updates that mix pretraining distributions (PPO-ptx).

For readers seeking to deeply understand the foundational technologies of current large model alignment, this classic 2022 paper, "Training language models to follow instructions with human feedback," is essential reading, recommended starting from the introduction and methods sections.

Fontes:《Training language models to follow instructions with human feedback》(2022)

Pesquisa

Risk Revisited: Why Volatility Is Not Risk, and the Future Is Unpredictable but Manageable

Verificado 2026-11-16 21:46 GMT+8

Leitura de longo prazo · 《Risk Revisited》(2014)

In the 2014 memo "Risk Revisited," Howard Marks, founder of Oaktree Capital Management, reexamines the nature of investing. He first challenges the academic definition that equates "volatility" with "risk." Marks states: "Volatility is the academic’s choice for defining and measuring risk." He argues that most investors do not truly fear volatility but rather permanent loss (permanent loss). Volatility is merely temporary fluctuation; as long as an investor holds through it, recovery is often possible. Permanent loss, however, is irreversible and may result from investors selling during downturns due to lost conviction, time horizon constraints, or financial urgency, or from fundamental deterioration in the investment itself.

Secondly, Marks discusses the unknowability of the future. He emphasizes: "Given the near-infinite number of factors that influence the future, the great deal of randomness present, and the weakness of the linkages, it’s my solid belief that future events cannot be predicted with any consistency." Since the future cannot be precisely predicted, risk cannot be a priori quantified. However, this does not leave investors helpless. Marks proposes that while we cannot know exactly what will happen, we can perceive the range of outcomes and their likelihoods. Therefore, "the future should be viewed not as a fixed outcome that’s destined to happen and capable of being predicted, but as a range of possibilities and, hopefully on the basis of insight into their respective likelihoods, as a probability distribution."

For modern readers, this theory provides a clear mental framework. Consider an investor facing a high-volatility technology stock portfolio; traditional views might suggest reducing volatility to "manage risk." But according to Marks's logic, if the companies within the portfolio have sound fundamentals and the volatility stems solely from market sentiment, then such volatility is not true risk. Conversely, if a company faces value destruction due to technological obsolescence, even stable stock prices imply significant risk. Thus, key risk management lies not in smoothing curves but in identifying factors leading to permanent loss and constructing a probability distribution model capable of withstanding multiple possible scenarios.

"Risk Revisited" was written by Howard Marks in 2014. It is suitable for investors seeking to move beyond simple technical indicators to deeply understand the essence of risk. Readers are advised to start with chapters distinguishing "permanent loss" from "volatility" to develop probabilistic thinking in an uncertain future.

Fontes:《Risk Revisited》(2014)

Pesquisa

Você está em dia nesta visualização

Entrar / Cadastrar
LeituraPesquisaRadarFramework de investimento

Based on this argument, Stephan put forward two specific demands: First, Nvidia must commit to achieving 100% renewable energy across its supply chain and data investments by 2030, ceasing reliance on fossil fuels and nuclear power. Second, it must invest in new renewable energy capacity globally, particularly in key supplier regions like Taiwan and in areas where Nvidia invests heavily, such as Southeast Asia.

The boundaries of this accountability logic are also worth noting. In its 2026 Sustainability Report, Nvidia stated that it had matched 100% of its global electricity consumption with clean energy for the second consecutive year, and noted it is designing platforms like Vera Rubin to advance powerful computing alongside sustainability. Regarding chips flowing into Russian drones, Nvidia stated that the relevant products are consumer-grade systems intended for students, developers, and startups. Stephan’s argument positions Nvidia as a bottleneck node in the entire AI ecosystem, but there is debate over how much actual control chip suppliers have over how downstream customers deploy infrastructure—this is precisely the weakest point in his accountability framework.

Stephan stated that the Berlin protest is just part of Greenpeace’s new global Big Tech campaign. Just days after the GTC protest, Greenpeace activists joined the Stop Dirty Data Centres coalition, conducting two days of localized protests in Europe and around the world. He explicitly stated, 'Peaceful protest and nonviolent direct action are part of Greenpeace’s DNA.'

For the AI industry, the value of this interview lies not merely in recording another protest, but in presenting a specific accountability framework: when a company acts simultaneously as a technology supplier, financing driver, supply chain organizer, and public promise-maker, environmental groups choose to concentrate pressure on a single node rather than dispersing it across the entire industrial chain. This strategic choice itself offers a lens through which to understand issues of social acceptance regarding AI infrastructure.