Both Woodford and Hall are well-deserved prize recipients and I’ll go over the contributions that led to them receiving the award. What makes this Nobel special, and to self-promote Nominal News a bit, the award was given for something that I have extensively written about in numerous articles – the New Keynesian model.

Nominal News is an economics newsletter written by a PhD Economist that translates the latest economic research into clear, policy‑relevant insights on current issues. It is a reader-supported publication. Join 4,000 readers to stay-up-to-date with Nominal News directly in your inbox:

Subscribe now

Michael Woodford and Putting Together the New Keynesian Model

The main contribution by Michael Woodford highlighted by the Nobel is the formalization of the New Keynesian (NK) model.

The Problem – How to Properly Model Inflation, Interest Rates and Unemployment

Macroeconomic models can suffer from a problem called the Lucas Critique. Suppose you want to model how inflation reacts to interest rates. What you could do is look at historic data of inflation and interest rates, and specify a relationship between them. For example, based on such an analysis, you could determine if you raise interest rates by 1 percentage point, inflation will fall by 0.5 percentage points.

The problem with this approach is that we cannot be confident that the historic relationship between inflation and interest rates will continue to be true in the future. That’s because other factors, such as new policies or new technological developments, can alter this relationship. This issue was stated by Robert Lucas, and is now known as the Lucas Critique.

To avoid the Lucas Critique issue, we need to microfound the macromodel – that is to model the economic behaviors of people, firms and governments directly. By modeling how these economic actors make decisions, if there is a policy change or new technology, we can see exactly how these economic actors will respond. We no longer need to rely on historic patterns. This is what the New Keynesian model proposed by Woodford did (and also why the prize description mentioned “...microeconomic foundations…”).

We have discussed the NK model extensively, as it was extremely effective in guiding us through the pandemic economic disruptions. The key microfounded components of the NK model are that prices are set by firms based on the following three factors:

• Real marginal cost (the additional cost to produce one unit of good)

• Desired mark-ups (the profit firms want to earn)

• Expected inflation (what firms believe inflation will be in the future).

Expected inflation is an important feature of the NK model and it comes from a microfounded assumption – ‘sticky prices’. Firms (or workers) do not change prices (wages) every day, but only occasionally. Therefore, when they do change the price, firms need to make an assumption of what inflation will be in the near future, because they will be locked in to the price they set for a period of time.

This ‘simple’ sounding model started quite a revolution in economics, and many more complex versions of this model have been made.

A Short History

The NK model was not a flash in the pan, eureka moment. It took years of previous economics research. I, myself, do not know the entire history, but I will rely on Professor Erik Oberg’s lecture slide to show the amount of research that led to Michael Woodford writing down the formal model.

Forward Guidance

As an aside, one thing that I found a bit amusing about awarding the Nobel to Woodford now is the contrast between his view on forward guidance and the view espoused by current Federal Reserve Chair Kevin Warsh.

Forward guidance occurs when central banks telegraph their long run goals and actions. This can be very specific, including pointing out under what conditions a central bank will raise/reduce interest rates. Woodford, in a 2013 conference paper, said this about forward guidance:

“Inflation-targeting central banks have been notable for the amount of information that they provide to the public, not only about their longer-run goals, but about the way in which they expect to conduct policy in the future in order to achieve those goals. For reasons discussed above, I regard this development as a positive one. Greater clarity within the policy committee itself about the way in which policy is expected to be conducted in the future is likely to lead to more coherent policy decisions, and greater clarity on the part of the public as to how policy will be conducted is likely to improve the degree to which the central bank can count on achieving the effects that it intends through its policy. “

On the other hand, Kevin Warsh in August 28, 2026:

“To highlight one example, I have set out to change the form and function of the Fed Chairman’s so-called forward guidance. You might know about my long-time discomfort with early pronouncements of future policy decisions. I much prefer another path . . . and will make the case for it.”

The fact that Woodford just received the Nobel with an opinion that is in stark contrast to the current Federal Reserve Chair does seem like an amusing coincidence.

An Important Contribution

The Nobel to Michael Woodford is not without its critics. Quite a few economists believe that Woodford deserves the Nobel for his overall body of work, but that the NK model is not good and shouldn’t have been awarded the Nobel. I think the reason for such a view is that even among economists, there is a bit of factionalism and preference for one type of view of monetary policy and inflation.

The NK model challenged some of the previous theories on inflation – such as inflation being driven solely by monetary policy or by government debt issuance (“money printing”) – which may have not made everyone happy. Derivatives of the NK model showed us how inflation can exist absent any actual money at all.

My belief is that the NK model was crucial in guiding the US economy through the turbulent Covid period. Thanks to the NK logic, the US was able to prevent any major recessions and rebound very quickly as one of the best performing economies post Covid. And, yes, it did come with inflation, which ‘unfortunately’ was the cost of the policy. However, the alternative – a stricter central bank which would have kept inflation in check – would have resulted in a worse outcome for all.

Robert E. Hall – Figuring Out People’s Behavior

The main paper by Hall related to the Nobel prize focused on people’s consumption patterns.

In 1957, Milton Friedman proposed the permanent income hypothesis (PH), which posited that people will base their spending decisions in a given year by taking into account how much they will earn throughout their life. A person’s lifetime average annual income is referred to as ‘permanent income’. From this permanent income, people choose to spend a fixed proportion, leading to a smooth consumption profile each year.

Hall built upon this theoretical idea and demonstrated that people do follow this behavior when making consumption decisions. Hall showed three things:

• Today’s consumption level of a person is the best guess at tomorrow’s consumption level;

• Past consumption is not useful in forecasting tomorrow’s consumption (the past does not matter):

• If a person has a wealth shock (for example, their stock market investments’ increase/decrease in value), people slowly adjust their spending over time to these shocks.

Hall’s key point was to show how much a person spends today is already influenced by what the person believes will happen to their income and wealth in the future. Why is this important?

Rational Expectations and Why Stimulating the Economy May Be Hard

Hall’s contributions gave us two things:

1. That people act by taking into account their beliefs about the future, i.e. rational expectations, and;

2. How to incorporate these rational expectations into a model.

This was then further incorporated by Woodford in the NK model, which is the link between the two Nobel prizes.

Another important implication of Hall’s work is about the impact of economic stimuli.

Spending matters for the economy – too much spending can lead to inflation, whereas insufficient spending can lead to a recession. During downturns, governments may want to boost the economy by giving people money.

But, if a person spends a fixed amount based on what they believe will be their lifetime average income, a one-off government payment should not alter a person’s spending. Receiving a single $2,000 check does not meaningfully impact a person’s lifetime income, and, thus, spending shouldn’t change. This means that the government might be unable to stimulate the economy when it needs to.

Foundations of Economic Research

The work by Woodford and Hall laid the foundations of a wide body of economic research. This Nobel is very well deserved, as their contributions have become an ubiquitous part of economics. Many research papers have extensively built on their work – some of which we have talked about before, and, which itself might win a Nobel prize in the future.

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe now

Share

Subscribe now

The US has recently issued several policy proposals that increase costs of finding a job in the US for foreigners and, especially, foreign students:

• A $70,000 to $130,000 fee on students planning to use OPT (Optional Practical Training)

• A $100,000 fee for a skilled immigrant (H-1B visa)

The OPT program allows international students to work for 1 year in the US, and up to 3 years if their degree is classified as science, technology, engineering or math (STEM). While international students are under OPT at a company, the company may apply for the H-1B visa on their behalf in case the company wants to keep the international worker.

These additional fees on both H-1B visa and OPT programs will heavily impact employment of foreign students in the US by making it significantly more costly. For context, the median H-1B employee earns around $130,000.

So what are the likely impacts of these two policies on the US economy?

Overall Economic Weakening

We have previously discussed the impact of restricting skilled immigration. The US already curtails skilled immigration by having a very limited number of H-1B visas available. Using the fact that currently all new H-1B employees receive the visa via lottery (at random), economists have extensively studied the impact on firms of winning and losing these lotteries.

Read more

Subscribe now

Podcast

Below is the link to our companion podcast for this article:

Countries differ by their level of inequality - below is map of countries by Gini coefficient (a measure of inequality):

The dark blue countries have low levels of inequality, while dark red have high levels of inequality.

The Nordic countries are often presented as paragons of economic equality. Why they have less inequality is an open question – one suggestion is that the Nordic countries are altruistic societies. But what if it’s the opposite – people in these countries are more selfish?

Selfish Equality

Stostad (2026) investigated the idea that selfishness may matter for lower inequality in his recent job market paper. Stostad ran two laboratory experiments on volunteers from different countries. In these experiments, he asked participants to play a game.

The Dictator Game

A participant – referred to as the dictator – was told that they are splitting an amount of money between themselves and a recipient. Initially, the dictator was presented with one of two games:

• An equal split game – $800 to the dictator and $800 to the recipient;

• An unequal split game – $1000 to the dictator, $600 to the recipient.

The dictator is then given a choice to transfer some of their money to the recipient – either $0, $100 or $200. The dictator is told that the recipient will receive whatever the dictator decides.

In a slightly different version of the above games, the dictator is told that the recipient will see the dictator’s decision and have a decision of their own to make – the recipient can choose to take away $600 from the dictator, and in return, the recipient receives $100.1

Thus, in this version of the game, as the dictator, you are now worried about what the recipient may decide to do. It might be prudent to give the recipient some extra money, if it reduces the probability that the recipient will choose to ‘take’ away the dictator’s $600.

Importance of Personal Beliefs

If you are the dictator, you are trying to predict how the recipient will react. If I don’t offer them a transfer at all, what is the probability they will take away $600? Similarly, if I give them a $200 transfer, will that probability decrease and by how much?

Stostad asked the participants specifically what they believed the recipients would decide in response to their transfer offers. Specifically, Stostad asked the dictators how many recipients they think will take $600, if, as the dictator, they transfer $0 or $200.

The purpose of this question is to elicit what the dictator believes about recipients. As an example, suppose the dictator believes that if they transfer $0, 60% of recipients will take their $600, but if they transfer $200, then only 40% of recipients will take their $600. Based on this, it means the dictator expects to see a benefit of (60%-40%)*600 = $120 by giving the recipient a $200 transfer.2

The purpose of estimating this dollar benefit is to see how much the dictator expects to receive in exchange for offering the transfer. This number is important as it suggests a non-altruistic motive to the money transfer – if you expect a larger benefit, you may be more likely to offer the recipient a transfer.

Behavior of Non-Altruists

Stostad wanted to look, explicitly, at the behavior of non-altruists in these games.3 Non-altruists are defined as individuals who do not transfer any money to the recipients in the first dictator game – the game where the recipient has no choice but to accept whatever the dictator decides. In this dictator game, if you choose to transfer, as there’s no financial benefit to you, you may be doing it due to altruism.

By focusing on non-altruists from this first game, we can claim that if you choose to transfer money to the recipient in the second version of the game, you might be doing it out of ‘selfish’ reasons rather than altruism. That’s because by transferring, you’re hoping to dissuade the recipient from taking your $600. These non-altruists are referred to as ‘self-preserving redistributors’ – they re-distribute (i.e. transfer) out of selfish reasons.

Country Differences

Stostad looked at how people in different countries responded to this experiment and categorized 4 types of respondents:

• Altruists – people who transfer even though there is no benefit to them;

• Self-preserving redistributors – people who only transfer when there’s a benefit to them;

• Never-redistributors – people who never transfer money to others

• Others – individuals who do not fit the above patterns.

The distribution of these individuals between countries is shown below:

If we focus solely on non-altruists, the percentage of non-altruists that are self-preserving redistributors are shown below:

This result gets quite interesting when we look at the relationship between the self-preserving redistributors (x-axis) compared to how much each country transfers via government programs to the bottom 50% (y-axis):

Countries that re-distribute more appear to have a larger share of self-preserving redistributors.

Moreover, if a country has more self-preserving redistributors, these self-preserving redistributors also believe that they will benefit more from transferring. The chart below shows the share of self-preserving redistributors (x-axis) against what they believe they will benefit from transferring (y-axis):

Lower Inequality Driven By More Selfishness

The above results suggest that people in certain countries may view inequality/unfairness as costly to them, because they may be ‘punished’ by recipients. They are more motivated to transfer money, to reduce the chances of punishment. These transfers are not motivated by any altruistic reason, but rather by purely selfish reasons motivated by one’s own beliefs on how recipients will react.

Stostad, with this paper, brings up a novel explanation of what might explain the differences in inequality levels between countries. It is not altruism, culture, history, etc., but, perhaps, selfishness and the beliefs about the costs of inequality. Naturally, this paper is not proof of this theory – Stostad is only presenting a correlation between the shares of self-preserving redistributors and the government transfer policies of countries. Moreover, the experiment itself might be somewhat limited, as people in experiments might not perfectly reflect real life behaviors.

However, this study opens up a new avenue of research and policy options. If inequality really is driven by people’s beliefs about the costs of inequality to them, then understanding and communicating the true costs of inequality to people becomes critical. Reducing inequality is not free – transfers need to be funded by tax revenues which have economic costs. But if inequality itself has costs to members of society such as increased crime, higher social unrest and lower social trust, then there might be an optimal amount of redistribution needed to reduce these costs.

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe now

Share

The dark blue countries have low levels of inequality, while dark red have high levels of inequality.

The Nordic countries are often presented as paragons of economic equality. Why they have less inequality is an open question – one suggestion is that the Nordic countries are altruistic societies. But what if it’s the opposite – people in these countries are more selfish?

The first part of the podcast is available as a preview for all subscribers and the entire podcast is available for paid subscribers. This week (October 16) is the last week of our 40% off Launch Discount -

Subscribe now

Refer Nominal News

This is the companion podcast episode to the following article:

Read more

The first part of the podcast is available as a preview for all subscribers and the entire podcast is available for paid subscribers.

This is the companion podcast episode to the following article:

Subscribe now

Refer Nominal News

Read more

Subscribe now

The companion podcast to this article will release later this week!

Smartphones have become ubiquitous. This has led to regular news stories blaming them for everything – birthrates, literacy, anxiety. Bad school behavior and poor test score performance is also blamed on smartphone use. Naturally, this a testable theory, and a large study by Allcott1, Baron, Dee, Duckworth, Gentzkow and Jacob (2026) (“ABDDGJ”) looked at what happened when phone use in school was restricted.

Reducing Phone Usage

Over 4000 schools in the US implemented a policy where smartphones needed to be put into Yondr pouches at the beginning of the day. These pouches are magnetically locked and prevent the use of the smartphone:

ABDGGJ compared schools that used the smartphone pouches with otherwise similar schools, and observed them for 3 years. So what do they find?

Impact of Smartphones on School Outcomes

Nothing. The authors found no difference in test scores, attendance, disciplinary actions or online bullying.

To be a bit more precise – there were relatively minor differences in outcomes:

• Students subjective well-being falls in the first year but turns positive in year 3;

• Disciplinary actions rise in the first year of smartphone pouch adoption but revert to pre-pouch adoption levels by year 3;

• High school students saw a slight improvement in test scores – a 0.9 percentile score increase;

• Middle school students saw a slight reduction in test scores;

To give additional context, a 0.9 percentile increase in test score is approximately one fifth of the gains students get when paired with a better teacher.

Potential Explanations

ABDGGJ posit a few explanations of the puzzling results. If the problem with smartphones is that they distract students, then by removing smartphones, students may resort to other distractions such as talking in class. So test scores are unlikely to change if students don’t substitute phone use with in class learning, but rather, with other distractions.

Some of the other findings have intuitive explanations:

• The rise in disciplinary actions initially may be driven by the fact that there may have been resistance to the smartphone pouch policy;

• Improved subjective well-being has been shown to occur with decreased social media usage in prior studies;

• The reason middle school students may have seen a test score decline, while high school students saw a test score increase could be because younger children have less impulse control, resulting in more disruptive behavior compared to older children.

Broader Research

Similar results have been found by other studies. Figlio and Ozek (2025) looked at the impact of smartphone restrictions in schools in Florida. Using phone signal tracking data, Figlio and Ozek were able to differentiate schools by smartphone usage prior to the ban, with the expectation being that high smartphone-use schools would see bigger improvements in test scores.

Overall, Figlio and Ozek found that the smartphone ban improved test scores by about 0.6 to 1.4 percentile, in line with the findings of ABDDGJ. Beland and Murphy (2016) conducted a study on UK students during the time period of 2000-2013 and found that there was a positive impact, similar in size to ABDDGJ. At the same time, some studies showed truly zero impact of phone bans; Kessel, Hardardottir and Tyrefors (2020) found that imposing a ban in Sweden did not change test scores.

Banning Smartphones Will Not Do Much

It appears that banning smartphones in school, at best, will only minimally improve test scores. Smartphones feel like an easy scapegoat for all educational ills, but research has failed to show that they are the cause.

However, if smartphones are an issue, the problem might be that they’re already too ubiquitous. Students use phones outside of school and the majority of any adverse impacts may be occurring in these settings, making school bans ineffective.

For example, a paper by Barwick, Chen, and Li (2025) looked at how phone app usage can be ‘contagious’. In a college setting, if Person A has a roommate that uses phone apps more than the average user (average phone app use is 92h a month with a standard deviation of 108 hours), Person A then uses phone apps more themselves. The authors also found that starting wages upon graduation drop by 2.3% if one uses apps one standard deviation more than average (i.e. 200 hours a month).

This study suggests the ubiquity of smartphone use might be the issue. Moreover, we also have research that shows that social media has adverse impacts on a person’s well-being (covered by us here). Therefore, the fact that smartphones are omnipresent may be the issue and simple school bans are unlikely to have an impact. Since smartphones will not disappear, some new approach may need to be developed to better deal with smartphone use.

One such idea could be based on the approach used by some schools, which have implemented training and courses on the healthy use of smartphones and social media. The Florida schools in the Figlio and Ozek study had such classes. Currently, school-based smartphone/social media instruction is minimal – only around 10% to 20% of students in the US had a standalone course offered. Perhaps a more structured class would result in a better relationship with smartphones/social media. After all, physical education and health classes are often year round courses, so a dedicated course on smartphones which is, ultimately, a mental health class may be warranted.

Let me know your thoughts/comments on smartphone use in school in the comment section!

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe now

Share

However, the exact cause of the college wage premium is hotly debated. Two main theories for the college wage premium have been proposed:

1. Completing college is a ‘signal’ that a person is highly capable;

2. College improves one’s abilities, making them more productive.

Find out in today’s podcast how we can test these theories and which one is the more likely explanation.

And if you’d like to read more about the topic, follow the link below :

Subscribe now

NEW! Podcast

Note to subscribers: we have officially launched a companion audio/podcast to our articles, which you can click to listen below. The companion audio talks about the topic I wrote about below, covering the article itself, but also discussing additional research that did not make it into the piece, as well as my opinion on the issue. Future companion audio/podcasts will be for Paid Subscribers, but is free for this article. We’re still running our 40% off launch special for a Paid Subscription!

College education is known to increase wages and earnings. The college wage premium – how much more one earns with a college degree over the earnings of someone with only a high-school degree – is quite large at 70%-80%, as shown here by Bengali, Sander, Valletta and Zhao (2023).

However, the exact cause of the college wage premium is hotly debated. Two main theories for the college wage premium have been proposed:

1. Completing college is a ‘signal’ that a person is highly capable;

2. College improves one’s abilities, making them more productive.

Waste of Time or Upskilling?

Under the ‘signalling’ theory, getting a college degree does not increase skills. However, since getting a college degree might not be easy (entrance exams and completing all coursework requires work), a person that successfully gets a degree can be assumed to have more ‘ability’ and therefore is more productive. If the person is more productive, firms are more interested in hiring this person and are also willing to pay a higher wage to this person.

Thus, under the ‘signalling’ theory, college merely serves as proof of one’s abilities that they already had. If there was another way we could observe a person’s ability, then college would become obsolete.

Contrary to ‘signalling’, the second theory proposes that college improves a person’s ability. Thus, after completing the degree, the person is more capable and more productive. Firms notice this higher productivity, and pay workers with college degrees more. Without a college degree, the worker would have a lower ability and thus, would not be as productive for the firm, which in turn means the firm would pay the worker less.

Changing College Requirements

Determining which theory is more likely to be correct is not easy. The reason for this is that we cannot observe the same person both getting a college degree and not getting one to determine how their wage would have developed. Occasionally, economists get lucky when a ‘natural’ experiment occurs.1

Arteaga (2018) used a change in the curriculum at Universidad de Los Andes, the top university in Colombia. In 2006, Los Andes decreased the required number of courses required to receive an economics degree by 12, and a business degree by 6. Additionally, the duration of college fell from 4.5 years to 4 years.

If after this change, there is no impact on wages, then it suggests that college acts more as a ‘signal’ (or at least additional coursework has limited value). But before we reveal the answer, let’s discuss whether this ‘natural’ experiment is valid.

The Experiment

The main idea behind the experiment is to compare the wages of graduates who had to do more coursework to get their degree, to individuals who had to do less coursework after the 2006 policy change. In order for this to be an appropriate comparison, the pre-2006 students and the post-2006 students have to be similar. For example, the change in policy should not result in students with very different abilities applying to Los Andes. If that were the case, then the comparison would be useless, as you would be comparing apples to oranges.

Arteaga looked at the characteristics of the Los Andes cohorts pre and post reform. There were:

• No differences in the number of students entering the economics and business programs;

• No difference in the test scores between the pre and post cohorts;

• No difference in the graduate rates.

The first two bullet points suggest that the ability of students entering Los Andes did not change after the reform. The last bullet point tells us that the same number of students graduated after the reform. The reason this is important is if there were an increase/decrease in the number of graduating students, the signal of being a graduate might change (for example, if more students graduate, then the signal of graduating might be weaker).

Given there are no differences in cohorts, any changes in wages after graduation can be attributed to the fact that there is less learning at Los Andes.

Results

In order to control for any wider market changes in wages, Arteaga compared the wages of Los Andes economics and business students to the wages of graduates from other top 10 Colombian institutions. The results were striking:

The chart shows the mean wages of each cohort. Prior to the change in policy (red line), wages for economics graduates from Los Andes were 36% higher than from other top 10 schools, while for business graduates, they were 50% higher. Post-reform, the Los Andes wages fell by 16% for economics students and 13% for business students.

One of my first concerns about the result was that, maybe due to the fact that the program is a semester shorter, the graduating cohort is younger and firms may value older people with higher life experience. Arteaga, however, ran the analysis controlling for age and found no change in the ultimate result.

Lastly, maybe there were some other changes at Los Andes that resulted in this dramatic wage fall. Arteaga conducted an identical analysis for law students at Los Andes, where there was no curriculum change. As expected, the wage for Los Andes law students did not change. This all suggests that the main reason for the fall in wages is that students took fewer courses.

Why the Fall

So how exactly do employers observe that new graduates have ‘less knowledge’? Arteaga conducted interviews with several of the top employers. Arteaga found that for many employers, the course content that was cut out of the economics program was crucial to perform well in the job. Recruiters, via tests, were able to figure the new graduates did not have the necessary skills. For example, the probability of being hired by the Central Bank of Colombia for recent Los Andes graduates fell by 16 percentage points. This suggests that students started performing worse in the recruitment process, meaning they had a reduced chance of getting a high paying job.

Moreover, in the field of economics, where knowledge is easier to test, the fall in average wages after the reform was instantaneous. The first ‘new’ cohort immediately saw a reduction in wages. On the other hand, in business, the fall in wages took several cohorts. This is because business hiring is less test heavy – businesses generally hire people from various backgrounds. Thus, it took several years of observing the new cohorts to realize that the new cohorts from Los Andes may have less training.

Signalling vs Upskilling

The Arteaga paper strongly supports the idea that the premium from higher education does come from gaining more skills. However, this does not necessarily show that there is also no signalling effect. Arteaga mentions that other research on Colombian universities has shown that signalling matters.2 Separately, Lange (2007) argues that signalling can account for at most 25% of the gains from additional schooling. This is because employers quickly learn whether the person they hired is good or not so good. Thus, a person that is not as productive, but managed to get into a good university, would be quickly found out by the firm.

It is important to note that the results found in this particular study might not necessarily replicate in other cases, such as when other majors are analyzed or if the universities are outside of Colombia. But one thing that the Arteaga research does show is that the additional years of schooling at college do generate value and maybe the recent increase in talk around college not being worth it is a bit unfounded.

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe now

Share

Subscribe now

AI continues to make news headlines – whether it is solving math problems or hacking companies. However, its benefit in the workplace is still up for debate. From a big-picture perspective, US productivity (how much output we produce per unit of input) measures provided by the Bureau of Labor Statistics (BLS) have not really increased1 in response to AI usage. But how about from a firm-perspective – a recent research paper looked at how AI use at work impacts employees – and the results have significant conclusions for how businesses and individuals should use AI.

An AI Experiment on Lawyers

Autor et al. (2026) ran a three month field experiment on 133 patent lawyers in the US. The lawyers were split into two groups – one set of lawyers (let’s call them the AI Group) was allowed to use a non-public AI tool as part of their daily work of drafting patent documents, while another group did not have access to the tool. Autor et al. then conducted two exams on these groups.

One test that was conducted twice – once after 10 days and once after 90 days – asked the lawyers in each group to draft a patent-related document. The AI Group could draft the document using the AI tool, while the non-AI Group could not use the tool. The second test, conducted after 90 days, had each group conduct a ‘redline’ exercise on a draft patent document.

In this redline exercise, lawyers in both groups were asked to review and comment on the draft document, without the use of AI. All the tests were evaluated by professional patent lawyers who provided a grade on the following 5 elements – “enforceability, accuracy, strategic ambiguity, completeness and clarity”. Each test was graded by two lawyers.

So what happened?

Better Work Product, But Mixed Experience Gain

The initial results were interesting:

• On average, lawyers that had access to the AI tool drafted better documents; but,

• On average, they did not get better at reviewing (redline exercise).

For the document drafting test, the output quality, on average, was at the 50th percentile for non-AI Group versus, approximately, the 65th percentile for the AI Group (so an average lawyer became an above average lawyer with the use of AI).

However, the average outcomes masked significant variance:

• Junior lawyers (less than 7 years experience) benefited from the AI tool when drafting, but;

• On the redline exercise, where AI was not allowed in both groups, junior lawyers in the AI group, showed no improvement in reviewing over junior lawyers in the non-AI group;

• Senior lawyers (> 7 years experience) in the two groups showed limited difference in quality of drafting, but;

• Senior lawyers in the AI group were better at redlining and outperformed their non-AI colleagues on the redline exercise.

Although AI allowed junior lawyers to create better drafts, it did not necessarily mean that they gained skills. On the other hand, senior lawyers, thanks to AI, appear to have gained experience.

More Equipped Individuals Gain More

Autor et al. investigated what was driving this dispersion. Two things stood out:

1. Junior performance in the AI group became more dispersed – among the junior lawyers who used AI, more junior lawyers got very low scores (bottom 20th percentile), but also more got above average scores (60th to 80th percentile) on the redlining exercise with fewer average scores than the non-AI group;

2. Senior lawyers in the AI group performed better on the redlining exercise by focusing on more substantive issues in the patent document review – this included focusing on maximizing a patent’s commercial scope, addressing legal liabilities and referencing legal principles as part of their review, instead of stylistic and administrative adjustments.

Senior lawyers appear to have used their prior experience to focus on more critical issues, instead of more trivial issues, which junior lawyers in general tend to focus on. Senior lawyers were also likely to fully re-write sections to improve the patent document strength, whereas junior lawyers only jotted down broad annotations. Overall, this suggested that prior experience and knowledge mattered on who benefited from AI.

My Takeaways – How to Implement AI at Work

The Autor et al. paper is a rare example of a study that explicitly aimed to measure the influence of AI in a non-routine job. Given the varied outcomes, AI can be a tool that, if used well, can enhance a business, but, if used poorly, can perhaps weaken knowledge acquisition.

Junior employees may create better quality outputs thanks to AI, but, as currently implemented, AI does not necessarily increase the skills and knowledge of all junior employees. In fact, AI might mask low performance or a poor employee-job match. Some junior lawyers drafted higher quality documents because of AI, but may actually have become worse lawyers due to its use (note: whether AI ‘caused’ these junior lawyers to perform worse is not demonstrated).

Seniors, who already have significant knowledge and experience may benefit from AI, as they are able to continue furthering their knowledge via the regular use of this tool. This shows that certain foundational knowledge is critical to get the most of AI. Naturally, senior employees are already pre-selected in a way – you likely wouldn’t become a senior employee unless you demonstrated a certain skill level in the first place.

So, for me, the takeaways from this paper are:

• Allow senior employees to use AI for work;

• Ask senior employees to share how they use AI with junior employees;

• Perform knowledge/skill assessments for junior employees to incentivize the gaining of skills, rather than just using AI.

The last point is perhaps the most important – the Autor et al. study shows that there is a difference between doing and understanding a job. With the rise of AI use in the workplace, really understanding the job appears to be an important skill differentiator.

What You May Have Missed From Nominal News

On September 16 – the US Federal Reserve hiked interest rates. Find out why this happened, why the Federal Reserve is signalling further hikes and what I think will be the trajectory of inflation.

As this was a paid article – I am running a 40% off yearly subscriptions till the end of the week. If you’d like to see what is available at the paid tier, I have unlocked the below link:

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe now

Share

Subscribe now

Summary

• The FOMC (the interest rate setting committee) decided by a unanimous vote (12-0) to raise interest rates to a range between 3.75% and 4.00%.

• Most FOMC voters expect to hike interest rates at least one more time this year.

• Of concern, Warsh argues that inflation is driven by the fact that the US economy has ‘strengthened’, although data suggest otherwise.

Commentary

The Statement

Below is how the FOMC statement has changed from the previous FOMC meeting:

The interest rate hike was expected by markets and is most likely the right decision, as I argued here. Perhaps somewhat surprisingly, there were no dissents, either for holding rates steady or for a further interest rate hike. In addition to the statement above, the Federal Reserve has also published what FOMC participants forecast macroeconomic variables to be in the following years - GDP, unemployment, inflation, and interest rates.

We are currently offering a 40% discount off of all subscriptions!

Interpreting the Statement and Decision

As we mentioned in our previous piece, inflation is running hot. As the Core PCE chart below shows, inflation has remained above 3% since the beginning of the year and accelerated as the year progressed.

The FOMC statement contains two noteworthy changes:

1. The statement no longer references supply shocks or energy.

2. The statement notes that the committee wants a “timelier return to the Committee’s 2% goal”.

The first point suggests that the FOMC is looking at multiple sources of inflation, not just energy inflation. As I have discussed, there are three factors that are pro-inflationary - tariffs, war and the suppression of immigration.

While the energy shock due to war is the largest of these, the other two are smaller, but more persistent. Nonetheless, the energy shock itself can also be expected to spill-over into broader price inflation - at some point, firms and workers will demand higher prices and wages to offset the loss in purchasing power caused by higher energy prices.

This removal of the reference to energy in the statement sounds much more ‘hawkish’ to me, especially when coupled with the second change - a ‘timelier’ return. It appears that the FOMC is concerned with inflation remaining elevated for too long and unanchoring inflation expectations.

Recall that the standard economic model of inflation - the New Keynesian model - links inflation to:

• Real marginal cost;

• Desired mark-ups;

• Inflation expectations.

Supply-side shocks push up the real marginal cost of production, which directly increases inflation. If these supply shocks are not temporary, there is a concern that this elevated real marginal cost will feed into inflation expectations, permanently raising the rate of inflation. The FOMC wants to avoid this, as re-anchoring expectations typically requires a stronger economic downturn.

The Federal Reserve Forecasts aka Dot Plot

The Federal Reserve also posted its forecasts for key macroeconomic variables. It is worth noting that, as promised, Kevin Warsh did not submit his expectations. Below is what participants believe Core PCE will be in 2026 and 2027 (the solid bar shows the number of participants who selected each value in September, while the dashed line shows their projections from June ):

Core PCE for 2026 is expected to remain elevated in the 3.3-3.4% range. I should add an important caveat - the Bureau of Economic Analysis (BEA), which calculates the PCE will slightly modify their data inputs, which is expected to mechanically reduce the current PCE readings by around 0.1 to 0.2 percentage points. It does not appear that the participants took this into account when forming their projections.

Below is the forecast for the interest rate:

Here is the same data, just in the more classical ‘dot plot’:

This is where things get interesting:

1. Most participants expect another hike in 2026!

2. Most participants expect high interest rates to remain elevated throughout all of 2027, with a real possibility of another rate hike in 2027.

I do not find this surprising - if we take a reasonable expectation for government policy in the next year, inflation will persist and we will need higher interest rates to bring it down.

One thing I do find puzzling in the forecasts is the disconnect between the inflation forecast for 2027 (at 2.5%) and the interest rate forecast (at over 4%). If inflation by the end of 2027 is at 2.5%, as measured by Core PCE, I believe rates will be lower - closer to 3.25%-3.5%. The disconnect is a bit puzzling to me.

Kevin Warsh Press Conference - Mostly Evasive Statements

The press conference by Federal Reserve Chair Kevin Warsh was, frankly, not very informative. In part, that’s what Warsh wants - he’s opposed to ‘forward guidance’ (explicitly stating future FOMC decisions). However, beyond avoiding forward guidance, Warsh was also cagey in explaining exactly why the current decision was made. Two things stood out to me - one encouraging and one concerning.

Oil Shock Comment

In response to the very question about the oil price shock stemming from the US-Iran conflict, Warsh mentioned that there is a danger that higher oil prices could spill-over to other prices. This is a real risk and it is reassuring to see the Fed taking it seriously. Here is a recent paper showing how supply shocks can feed into inflation expectations, much like they did during Covid.

The basic argument of the paper is that general awareness of supply shock price increases (like those we’ve seen with energy) makes consumers less sensitive to price increase in general. This allows firms not directly impacted by supply shocks to raise prices.

Overall Economy Comment

In response to a question on what had broadly changed since the July meeting to prompt the rate hike, Warsh mentioned the following things in this order:

1. A strengthening US Economy

2. Inflation trends that were not passing the test

3. Geopolitics

I find this response concerning.

First - the US economy is not actually strengthening. Real GDP growth slowed from 2.1% in Q1 2026 to 1.5% in Q2 . These growth rates are lower than real GDP growth in 2024 (2.8%) and also in 2025 (2.1%).

The unemployment rate is similar to in 2024 at 4.1% and minimally lower than in 2025 (4.4%). Productivity growth was also low: 0.8% in the first quarter, and 1.4% in the second quarter, which are both below the long run average of 2.1%. Lastly, real wage growth has also fallen and is hovering near 0%.

None of these data points suggest a ‘strengthened’ US economy, as Kevin Warsh claimed.

Regarding inflation trends - I am also a bit surprised. Since July, it has been quite clear that inflation prints would remain elevated due to government policy choices. The inflation trend should have been anticipated by Warsh.

On the last point, what stands out is that geopolitics, along with the wider government policy, wasn’t mentioned as the first reason! Let me elaborate

Government Policies

Here is a chart of Core PCE inflation (the FOMC’s preferred interest rate measure).

Core PCE was trending down and bottomed in April 2025 at 2.6%. Then came the first supply shock: the ‘liberation day’ tariffs enacted in April 2025. At the same time, the US government enacted a policy that significantly reduced all forms immigration, creating a second supply shock.

Then, in February 2026, the US struck Iran, which led to the oil shock - the third supply shock. In response to these three supply shocks, the FOMC in the period 2025 to 2026 did not raise interest rates (and actually cut interest rates twice near the end of 2025).

Because these supply shocks are all inflationary, while FOMC policy was accommodating, it shouldn’t be surprising that inflation started creeping back up. Deferring interest rate hikes could be justified if we believe the shocks are relatively short-lived. However, they clearly weren’t, justifying the hike today.

Where I Think Inflation and Interest Rates are Heading

With the above assessment in mind, I do not see any of these supply shock pressures subsiding. Tariff policy continues to be pushed and we haven’t seen all the tariff impacts on inflation (new autos is one such category). Immigration remains curtailed and the conflict in Iran appears to not be coming to a resolution.

These factors are all beyond the FOMC’s control, yet the Committee needs to react to them. One could argue that the FOMC should have increased interest rates by 50bps given they plan on increasing anyway. I wouldn’t have been surprised if at least one dissent would have been in favor of a double hike. My guess is that the Committee prefers a wait-and-see approach to how government policy develops.

Overall, my forecast is as follows:

• If we see a meaningful government policy shift (e.g. reopening of the Hormuz strait), rates will be kept as they are now, with inflation expected to gradually fall by early 2027.

• If current policies remain in place, one more interest rate hike, perhaps at the next meeting

• If there is an escalation in either tariffs or the war, expect the upcoming 2026 hike, plus an additional hike in 2027.

Given these scenarios depend on political variables, I’ll let you decide which outcome is most likely.

We just launched Paid Subscriptions now that I officially went full-time and am currently running a 40% discount launch special. Paid Subscribers will get ‘live’ reporting on economic events, audio discussions of articles and ad-hoc economic analyses.

Subscribe now

Corporate profits are rising to record levels in the US. Part of this increase can be attributed to Trump’s corporate tax cuts enacted as part of the Tax Cut and Jobs Act (TCJA) passed in 2017. Under this Act, corporate taxes were reduced from an approximately 35% tax rate to a flat 21% tax rate. Given this large corporate tax cut, many are proposing restoring the corporate tax rate. But is that a good idea?

Corporate Tax Cuts, Wages and Employment

Nearly all taxes are distortionary, as they change people’s behavior. Corporate taxes are no different, which is why Kennedy, Dobridge, Landefeld, and Mortenson (2026) (“KDLM”) looked at what the impact of the TCJA corporate tax cuts was on:

• Firms’ hiring decisions;

• Workers’ wages;

• Firm profits.

If you’d like to skip over the methods of the paper and go straight to the results, please scroll to section II.

I. Methods Section: How KDLM studied this question

To study the impact of corporate tax cuts, KDLM used corporation tax filings from 2013 to 2019. In addition to these filings, KDLM had access to employee tax filings at these specific corporations (this data is called Linked Employer-Employee Data). It’s worth emphasizing that KDLM did not use survey data or estimates, but actual tax filings, which allowed KDLM to precisely observe firm revenues, worker/owner incomes, firm expenditures, etc.

It might be tempting to assume that the next step would simply be to look at how firms’ tax filings changed from pre-TCJA cuts to post-TCJA cuts, but this would be incorrect. That’s because under such an analysis, you’d be capturing not only the change in corporate taxes, but also anything else that was changing during that time period (like GDP growth or other regulatory changes). Your result would be incorrect. So what did KDLM do to deal with this?

C-Corporations and S-Corporations

The US tax code is extremely complicated, but occasionally it works out great for economists. KDLM noted that there are two types of corporations in the US: C-Corps and S-Corps. The main tax difference between them is that owners of C-Corps are double taxed. First, the C-Corp pays a corporate tax and then, C-Corp owners have to pay a dividend tax if they take out money from the C-Corp. For owners of S-Corps, on the other hand, there is no corporate tax - all the income generated by an S-Corp flows to the owners who pay standard income tax (like workers).

In addition to the corporate tax cuts for C-Corps, the TCJA also cut income taxes (which benefited S-Corp owners) and also gave S-Corp owners one other tax benefit that allowed them to lower their taxes.1 The TCJA ended up reducing taxes for C-Corps and S-Corps but at different amounts; C-Corps saw tax rates drop from 35% to 21%, while S-Corps saw effective marginal tax rates drop from around 40% to 29%. The chart below shows how the marginal tax rate (left chart) changed between the two types of corporations, and how the observed (in the tax filings) marginal tax rate changed.

This tax code difference resulted in C-Corps paying 26% less tax relative to S-Corps, leading to a 5.5% higher rate of return for C-Corps. Using C-Corps and S-Corps that are similar in size and operate in similar industries, but saw different tax reductions, KDLM could estimate the exact causal impact of tax cuts.2

So what did they find?

II. Impact of Corporate Tax Cuts

For a 1% increase in the net-of-tax rate (this is the share of income left after tax or 1 minus the tax rate)3, C-Corps saw:

• The amount of capital (e.g. machines) increase by 0.82%;

• Payroll (i.e. total wages paid) increase by 0.37%;

• Employment increase by 0.23%;

• Taxable income increase by 0.7%.

This tells us that corporate taxes do distort business decisions - with lower corporate taxes firms invest more, hire more, pay more and generate more revenue/sales (taxable income is higher). (As an aside, this is the typical cost of taxation, referred to as deadweight loss of taxation).

Now although we see that reducing corporate taxes increases economic activity, who actually benefits from this increase?

First, after-tax profits rise by 1.06% for every net 1% increase in net-of-tax rate (for the tax cuts received by C-Corps, this would imply an approximately 23% increase in after-tax-profits). So we know that shareholders benefit, but do workers?

Worker Wages

Perhaps surprisingly (or unsurprisingly), only workers in the top 5% of the income distribution at each corporation benefited from the TCJA tax cuts. The 95th percentile worker saw 0.22% rise in wages for every 1% increase in net-of-tax rate. For reference, these are workers that earn on average $170,000 and saw a $2,100 increase in wages.

A bigger increase went to executives - the top 5 highest paid employees (earning an average $375,000) saw a 0.47% increase in wages for every 1% increase in net-of-tax-rate. This translated to approximately a $9,700 increase.

Could these executive wage increases be justified? A common argument made in favor of executives is that they are the decision makers, which leads to firm growth and profitability. KDLM looked at several potential variables - sales growth, profit growth and relative sales growth to competitors. KDLM did not find any of these variables to explain the executive wage increase, suggesting that executives were simply engaging in rent sharing (i.e. executives earned higher wages without a corresponding increase in business performance).

Distributional Impact

Lastly, KDLM looked at the full picture of the distributional impact of tax cuts.

• 60% of gains flow to owners,

• 8% to executives,

• 32% to high-paid workers (top 10% of national income distribution)

• 0% to low-paid workers (bottom 90% of national income distribution)

Since workers can have shares and be owners, KDLM also looked at how the gains were distributed by income (versus by occupation):

• The top 1% got 33% of the gains from the corporate tax cut;

• The 90th percentile to the 99th percentile got 55%;

• The bottom 90th percentile got 13%.

Thus, whether looking by role or by income bracket, the gains predominantly accrued to the top of the distribution.

Interestingly, KDLM also found that 20% of the corporate tax cut benefits went to foreigners (foreigners can hold shares in C-Corps). Within the US, the gains mostly accrued in the Northeast and West of the US (darker color implies bigger gain):

III. What To Do With Corporate Taxes

Corporate tax cut proponents will argue that these taxes distort the economy and reduce growth, which is true. Corporate tax cut critics will argue that they increase inequality, which is also true. This makes corporate taxes very controversial.

Prior to writing this article, I viewed corporate taxes as an inefficient tax. Since a corporation is not an actual person4, I always considered focusing on the taxation of people that benefit from corporations - owners (capital/dividend tax), executives and workers (labor tax), and consumers (sales tax). Moreover, corporate taxes also create a global ‘corporate-tax’ race with many countries offering a zero-corporate tax policy as an incentive to encourage corporations to move. This ends up creating plenty of bloat, where corporations work on optimizing for tax purposes, which, one can argue, does not create real value.

However, as the KDLM paper shows, cutting corporate taxes acts as a regressive tax, as the cuts benefit high income individuals far more than low income individuals. It is worth noting (and KDLM mention this) that we do not know about other benefits corporate tax cuts may have, such as firms offering lower prices, which benefits consumers. I am, however, skeptical that even accounting for these other benefits would offset the regressive impact of the tax.

So what should we do about corporate taxes? Raising them may be one option, but we do not know if we will observe the reverse effects of corporate tax cuts - that is, higher-income individuals will ‘pay’ most of the corporate tax increase. Alternatively, the corporate tax cut should be offset by accordingly raising other taxes that will target beneficiaries of the corporate tax cuts. This would entail raising capital/dividend taxes and raising the marginal tax rate for higher-income individuals.

So to offset the corporate tax cut (or any future corporate tax cut), the marginal tax rates (labor, capital) on its biggest beneficiaries, the top 10%, would need to be raised. This would shift more of the tax burden on the groups that gained the most from the tax cuts. However, if such taxes cannot be raised, then the issue of raising corporate taxes may be warranted.

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe now

Share

Nominal News - Recent Paid Article

On Friday, I posted my first paid article in anticipation of the FOMC meeting on September 16. Below is the article, which I have unlocked for all. My next paid article will cover the FOMC meeting this Wednesday - if you’d like to access it, please consider upgrading to paid.

Subscribe to Our Newsletter

This morning, the Bureau of Labor Statistics (BLS) released the Consumer Price Index (CPI) data. The headline CPI came in at 0.4% month-over-month, 3.4% year-over-year, while Core CPI (not including food and energy) came in at 0.3% month-over-month and 2.4% year-over-year.1

Another data release published on Thursday - the Producer Price Index (PPI, the price that sellers receive) - also came in hot. Headline PPI came in at 5.4% on a year-on-year basis, while Core PPI came in at 4.7% year-on-year.

Although CPI is not the main inflation measure used by the Federal Reserve - that is the Personal Consumption Expenditure Price Index (PCE) - the CPI and PPI data feed heavily into PCE. Based on the releases so far, forecasters are assuming a high print for August PCE as well.

Moreover, the latest PCE readings for the month of July were far above the Federal Reserve target of 2% - headline PCE was 3.7%, while Core PCE was 3.3%. With inflation being above target, the focus has turned to the interest rate decision that will be made on September 16, 2026 by the Federal Reserve. The big questions are whether the Federal Reserve will hike, and whether they should hike.

Below I lay out how to think through this decision using a standard economic model. I will also address why I think many opinions are off the mark in this current interest rate discussion:

• “It’s a supply shock, we shouldn’t hike into a supply shock”

• “Inflation is actually low and falling so we shouldn’t hike”

• “The labor market is weak and thus we shouldn’t hike”

Bringing in a Framework - the New Keynesian Model

To understand where inflation is heading in the next several months, we need to look at the workhorse economic model - the New Keynesian model. In this model, inflation is determined by three elements:

• real marginal cost;

• the level of desired mark-up; and,

• inflation expectations.

The key elements we need to worry about today are real marginal cost and inflation expectations.

Real Marginal Cost

Real marginal cost is the input cost for firms to make one additional unit of a good or service. Thus, anything that affects the supply of inputs will push real marginal cost up. The largest recent increase in real marginal cost was the oil price spike. But besides oil prices, there have also been other cost spikes:

• Tariffs;

• Reduction in immigration.

Both of the above government policies also increase the real marginal cost of producing an extra unit, albeit to a lesser degree than the oil spikes.

Since these costs have spiked, inflation in the US has accelerated since the beginning of the year:

What Brings Inflation Down

The Federal Reserve can respond to this increase in inflation by raising interest rates. The main way raising interest rates works in reducing inflation is by making it costlier for businesses to operate (raising the cost of capital). This in turn reduces employment, which reduces wages. Lower real wages directly reduce the real marginal cost, bringing inflation back down.

Raising interest rates negatively impacts the labor market, but what if the Federal Reserve doesn’t hike interest rates?

Inflation Expectations

The other key driver of inflation is inflation expectations - what market participants believe inflation will be. This is used by firms (and workers) for price setting (for wage negotiations). Inflation expectations can make inflation a self-fulfilling prophecy - if I believe inflation will be 5% next year, I set prices today 5% higher, making inflation next year equal to 5%. That’s why central banks focus on inflation expectations and want to keep them anchored to the target they set.

In our case, if the Federal Reserve doesn’t hike and lets inflation remain elevated, there is a worry that everyone will no longer believe the 2% target and assume a higher target. This un-anchoring is dangerous, as re-anchoring the economy to a 2% target may require much much larger interest rate hikes to convince market participants to change their inflation expectations.

Why I Think a Hike is Warranted

Using the above model, I see two main concerns that impact both real marginal cost and inflation expectations:

1. Higher inflation over the last 6 months impacting expectations;

2. The continuation of government policies that all increase the real marginal cost.

All the government policies that have increased inflation over the last half year are still present. The war in Iran is unlikely to end any time soon. Tariffs are being brought back. Immigration is being made more costly and more difficult. All of these policies only point in one direction - higher real marginal cost.

At the same time, markets are catching on to the idea that inflation will be higher for longer - yields on US Treasuries are at record highs suggesting that the markets expect the Federal Reserve to raise interest rates. The worry now is that these inflation expectations may have become entrenched.

As inflation has now been elevated for many months and primarily driven by government decisions, a lack of action on the part of the Federal Reserve may signal that the 2% target is fluid. This on its own warrants a hike.

But, moreover, if inflation remains elevated in the next few months, as I expect, the Federal Reserve will almost certainly need to hike later in the year. Given that there are larger risks to not hiking interest rates, I believe the interest rate hike might as well come in now.

So if I were a voting member of the FOMC (the committee that sets interest rates), pencil me in for a hike.

Answering Some Common Statements

“We shouldn’t hike interest rates into a supply shock”

There is no economic rule that says not to hike into a supply shock. The main reason why we would avoid hikes is if we believe the supply shock is temporary. For example, suppose a hurricane reduces oil shipments. We know the hurricane will end, and thus we do not need to change interest rate policy, but rather let inflation be elevated for a bit.

In the current case, all the supply shocks (war, tariffs, reduced immigration) are driven by government policies. These policies have been with us for several months with no plan to end them. As they are not temporary, it is not your typical supply shock.

“Inflation is actually low and falling, so we don’t need to hike”

It is true that core CPI has fallen. However,

1. The main inflation measure we use is PCE, which hasn’t fallen;

2. Energy shocks take time to feed into core CPI.

With Federal Reserve decisions, we need to be forward looking. As there is limited disinflation going forward, we should expect inflation to creep up, absent any Federal Reserve actions.

“The labor market is weak and thus we shouldn’t hike”

The labor market is mixed - with both strengths and weaknesses. Unemployment is low at 4.1%. Wage growth is a bit more mixed - certain estimates have it at 3.1% while others at 3.6% (for job stayers). This is basically zero real wage growth. The problem, however, is that inflation is still elevated and climbing. Absent government policy changes, the only way to bring inflation down is through further real wage deterioration. Unfortunately, workers often take the brunt of inflation reduction.

Subscribe to Our Newsletter

Hello,

Today, as the last part of the Nominal News re-launch, I am debuting a Paid Subscription Tier, in addition to the current Free Tier. I want to give you an update of what to expect, and explain what will be provided in the Paid Tier.

Paid Tier

The Paid Tier Subscription will, in addition to the weekly deep dives that are part of the Free Tier, provide the following:

• Special edition articles ~twice a month, covering breaking economic developments, as they occur;

• Audio coverage of certain deep dive articles to give further information and opinion on the topic covered;

• Early access to my own data analyses;

• Prioritizing responses to any questions about the weekly deep dives.

Special Edition Articles

In these articles, I will focus on applying academic economic research to breaking developments, as well as provide more of my opinion on the subject.

As an example of the type of article, over the last few weeks, I have provided 4 Special Edition Articles:

• Special Edition: Federal Reserve FOMC Meeting Update

• Special Edition: Responding to Stephen Miran’s Tariff Op-Ed

• Special Edition: A Bizarre Press Conference After the FOMC Holds Interest Rates Steady

• Special Edition – US Treasury Increases Bond Buybacks

As you can see above, the topics covered were interest rate announcement meetings, tariff arguments and the most recent US policy of expanded bond buybacks.

I will publish around 2 Special Edition articles per month for Paid Tier subscribers and, as relevant for major economic developments. The next one, for example, will cover the FOMC Meeting on interest rates on September 16.

Pricing

The paid Tier Subscription will cost $100 for a yearly subscription. But, as part of a launch promotion, I am offering a 40% discount – so $60 for the first year (which works out to be $5 a month). This new subscriber offer will only last through September 18.

I will also offer a monthly version of the Paid Tier that will cost $30 (discounted to $18 per month for the whole year, as part of the launch through September 18), in case you’d like to try it out first.

The reason why the stand-alone monthly subscription is higher than the annual is two-fold:

1. I believe the most value will be derived from being a subscriber for a year, as articles will connect thematically with each other, and;

2. By having longer-term subscribers, I will be able to focus my writing on topics that I deem important rather than focusing on topics that will retain short-term subscribers and get clicks.

Lastly, if you’d really like to champion my mission of sharing economic knowledge and insights with the wider community, please consider becoming a Founding Member at $1,000 per year. Your support will help accelerate Nominal News and allow me to enhance the offering, of which you will be the first to know! You will receive a personalized Nominal News stein mug as well.

Upgrade to Paid

Free Tier

Under the Free Tier, you will continue to receive weekly emails that will mostly focus on deeper dives into economics research and policy. There will be no change here and you will continue to receive weekly emails – most likely on Mondays. As one of my goals with Nominal News is to share my economics knowledge with the wider public, I would not want to paywall this.

I am already very grateful for your subscription, and if you’d like to support me in other ways – likes, comments and sharing the articles are always extremely helpful! Actually, I have enabled the Substack reward program, which will grant you a reward for referrals here:

• 5 referrals – 1 month Paid Subscription

• 20 referrals – 1 year Paid Subscription

• 50 referrals – a Nominal News themed stein mug (I use it for tea and beer!)

Thank You

As always, thank you very much for supporting my work and my mission. I would not be able to launch all this without your support! I am very grateful and excited to continue the Nominal News journey with you!

If you have any questions/comments, please do not hesitate to reach out on substack or via email, nominalnews at gmail.com

Now onto a quick topic:

II. Inflation Measures By Income Groups

Inspired by a piece written by Substack Writer Michelle Teheux on poverty lines and how a single measure fails to capture how people experience poverty, I went ahead and did my own analysis on how inflation impacts people in different income brackets.

It should not be surprising that a much richer person might have a different consumption basket than a person living at the poverty line. One could assume that a higher income person will more likely spend a higher share on various services, while a lower income individual might spend more on necessities such as food, rent and energy.

As an exercise, I would like to ask you to think about how a consumption basket between a top 5% income individual and bottom 20% individual might differ, and who experienced higher inflation in the last 12 months (and by how much).

Consumption Basket Differences

To understand the spending patterns by different income groups, I relied on the work of Zheli He, Xiaoyue Sun and Efraim Berkovich of the Penn Wharton Budget Model. From their analysis:

Lower income individuals (left most column) do spend a lot more on food, shelter and energy. The lowest income group spends around 61% of their income on these items, while the highest income group (right most column) spends only 48%.

Measuring Inflation

Since each income group has slightly different spending baskets, it should not be surprising that each group may experience inflation a bit differently. For example, if food inflation is higher, then lower income groups experience higher inflation.

To analyze this, I used the above the spending percentages1, and applied the actual recent inflation in each of the spending categories from the Bureau of Labor Statistics.2

Based on the spending patterns of each of the income group, I approximated the inflation rates faced by the lowest income group and highest income group over the last 12 months:

The inflation rate faced by high income individuals lately has been much lower than the officially stated rate. Moreover the gap itself is quite significant – almost a 0.6 percentage point difference in inflation rates.

Rethinking Real Wage Growth

The above analysis also points to a slightly different perspective on how much real wages grew for every household. The Federal Reserve of Atlanta’s Wage Growth Tracker shows that as of July 2026, the lowest income quartiles saw a 3.7% growth in wages, while the highest income quartile saw a 3.8% growth in income. With the above inflation measures, that means the lowest income households saw nearly no real wage growth in the last year, while the highest income households witnessed a 0.8% real wage growth.

This suggests that many in the lower quartile of income have seen their incomes stagnate over the last year. The adjustment for inflation by income could also adjust our view of what real wage growth workers saw since the pandemic. Post pandemic, low income households saw some of the strongest wage growth. However, it might have been muted by particularly strong inflation in categories they spend. This is something I’ll look into.

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe to Our Newsletter

Share

Subscribe to Our Newsletter

I have recently started writing full time and am turning on a paid feature in September, in addition to the current free offering. If you’d like to help me with some feedback (and a chance to win a 3-Month Paid Subscription tier) – please fill out the questionnaire below:

Nominal News Reader Questionnaire

Hello,

After going full-time on Nominal News earlier this month, I would like to introduce myself:

My name is Pawel and I am the writer behind Nominal News. To give a bit of background, I completed my PhD in Economics at New York University and live in New York City. My research at NYU focused on macroeconomics, trade and labor.

Specifically, my main research paper focused on the issue of how trade liberalization can unevenly impact people’s lifetime earnings, generating a situation where some people (in this case, it was manufacturing workers) are explicitly worse off under trade liberalization, even though the majority are better off.1 Here you can find my personal site with my research (I also hope to progress these papers as I have not had the opportunity to work on them since 2021!).

Why Nominal News

During my PhD, I had the privilege of hearing from many brilliant economists.2 Through these experiences, I got to learn the economic intuition and way of thinking that I realized is often restricted to the confines of academia. This inspired me to share this knowledge and experience with others, which is why I launched Nominal News.

With Nominal News, I want to teach you the economic tools and way of thinking that I learned during my time in PhD, rather than just focusing on giving ‘an answer’. Economic thinking can be used in nearly all settings, as, to me, economic thinking is modeling human behavior – how do people respond to things and to each other.

For example, suppose a government proposes a new tax-advantaged savings account for children. The initial thought might be that such an account will help people save on behalf of their children. But a deeper dive would show us that such an account is unlikely to generate new savings (wealthier people will save less in other accounts; lower income individuals often have insufficient income to save). This type of understanding can shine a completely different perspective on a policy and what it will actually achieve.

With Nominal News, I hope that this type of ‘deeper-dive’ thinking becomes second nature when considering economic policies,or any social policy.

Beyond Economics – How Nominal News Can Help Elsewhere

These tools and methods can be applied to anything that involves human behavior – economic policy, business decisions and even personal life choices. Some examples of questions that economic thinking may help:

• If I’m setting prices for products, what should they be going forward? – understanding how government policies impact inflation will be important

• Suppose I implement a new inventory or payments system, how can I determine if this system is generating benefits?3

• Should I pursue a very technical, narrow degree or go for a more broad liberal arts degree?

Economic thinking can help you with these questions and also ensure that you ask the right questions around these problems.

What To Expect

Going forward with Nominal News, I will continue to remain research-based. I will aim to share the methods and assumptions behind the research, and not just the results (as these results can be driven by the selected assumptions).

Any insights or opinion on my end will always be driven by research. I will also aim to keep the distinction between the research and my opinions separate. I think these are important tenets to be driven by, as a lot of public economic commentary can often seem like it’s based on the best available research, but really, it’s just a person’s personal policy preferences.

Article: The US Census – Hosting An Unnamed ‘Analysis’

On August 18, 2026, the US Census Bureau, a federal statistical agency, posted on its website, a report called: “Noncitizen Voting in the 2020 Election, A Beginning Analysis”.4 The report touches on a politically charged topic of whether ineligible people voted in US elections. But the report itself had two large red flags – it had no author and it lacked critical methodological elements.

The US Census Bureau

The US Census Bureau is a highly regarded institution. Its role in the US is critical, as the data it collects ends up being used to allocate seats across the US for the House of Representatives. It also has staff researchers that publish analyses based on the highly sensitive Census data. Every report typically has a specific author and a well described methodology section. Here is one such example “Monthly and Episodic Poverty: 2024” authored by Marissa Nicole Davila.

On the other hand, the Noncitizen Voting in the 2020 Election, A Beginning Analysis report contains no authorship nor any serious discussion of methodology. Both of these issues immediately raised concern in the economics community, as it suggested that the level of analysis is not something that would be considered acceptable.

Besides the lack of authorship, experts highlighted that the key term used in the report – non-citizen – is not actually a legal definition. This is a serious issue: in a February 2026 investigation of a Texas county, individuals were identified as non-citizens (and thus, the implication that they are not eligible to vote). However, it turned out that many identified individuals were citizens.5

The concerns raised by the research community turned out to be vindicated, as NPR found that the published report was not written by anyone at the US Census, but rather written by a politically linked think-tank:

“Work on the analysis was not conducted by career civil servants at the Census Bureau, and the team behind it included individuals affiliated with the America First Policy Institute, a think tank started by officials from Trump’s first administration, according to a Census Bureau employee who NPR has agreed not to name because they fear retaliation at work.”

The Importance of Institutions

The reason this issue is significant is that the US Census is a trusted source for rigorous data collection and research. It’s not a surprise that people would want to use the US Census as a seal of approval on their ‘research’. Such undermining of the institution can have long run consequences, as politicizing such institutions can only lead to bad outcomes.

It was promising to see that no one at the US Census put their names on this poor quality analysis. Moreover, the broad and swift criticism of this analysis is another good sign, as it helps protect the integrity of the institution. Regardless, it’s very unfortunate that such an analysis was published on their site.

Next week, I will officially launch a paid tier. My post next week will cover how it’ll work (Note: most importantly, weekly deep dives will remain free!)

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe to Our Newsletter

Share

Subscribe to Our Newsletter

Going forward, special edition articles will be accessible to paid tier subscribers. My next special edition article will be covering the FOMC meeting and press conference on September 16. Subscriptions will go live September 8. If you’d like to enter a chance for a free 3-month paid subscription, please fill out my questionnaire.

US Treasury Secretary Scott Bessent announced on August 19, 2026 that the US Treasury will boost buybacks of US treasury bonds with maturities between 10 years to 30 years. The US Treasury is expected to buy up to $4bln worth of bonds per transaction, 4 times a quarter. So what does it mean?

Rising Yields – i.e. Rising Interest Rates

The US has been experiencing a rise in yields on long-dated bonds. A yield can be thought of as the interest rate an investor would require to earn in order to buy the bond. Below is the yield of a 30-Year US Treasuries over the last 5 years:

Higher yields mean that when the government issues new 30 Year Treasuries (government debt), it will have to offer the yield as a fixed interest rate for the duration of the bond. This means issuing debt for the government has gotten more costly over the last 5 years.

At the same time, US Treasuries are often used as a baseline cost of debt for other markets. For example, the 10-Year Treasury is often relied upon to set interest rates for home mortgages. Other investments also rely on this as a benchmark.

This recent increase in yields and persistently higher yields appears to have concerned the government, which has been calling for lower interest rates for a while. The buyback initiated by the US Treasury appears to have been motivated by the desire to lower interest rates.

Buybacks – Balancing the Yield Curves

By buying back long-dated bonds, the yield on these bonds fall.1 In exchange, the US Treasury issues short-dated bonds (for example, 2-Year bonds), which have a lower yield than the long-dated bonds. This action does increase the yield of these short-dated bonds. In the end, the difference between yields of 2-Year bonds and 10-Year bonds shrinks.

Why Do This?

One potential reason to do such an operation is if the government believes the market is mis-pricing the bond. That is, if long-dated bonds have yields that are too high compared to what they think they should be, it might make sense to briefly issue more short-dated bonds, until the yields on long-dated bonds adjust.

However, this presumes that there’s a reasonable chance these long-dated yields will drop in the foreseeable future.

What Happened After The Buyback?

Right after the announcement, the 30-Year bond yields fell. But – yields quickly rebounded in a few days, as can be seen from the chart below (the red line being the day prior to the announcement)

What “Went Wrong”?

First, the amount of bonds being bought back is quite low, which is insufficient to truly move the market. Second, more importantly, the market disagrees with the US Treasury’s perception that the bonds are mis-priced. The yields, which are driven heavily by the Federal Reserve Interest Rate, are expected to be elevated until inflation is brought down. Basically, because inflation is expected to be higher for longer (due to multiple other government policies – tariffs, anti-immigration, war), interest rates are expected to be higher for longer as well.

In turn, investors demand higher interest rates (yields) to hold long-dated government debt. That’s because the yield of long-date bonds has to equal the yields of a sequence of short-dated bonds (the interest earned on a 30 Year bond should equal the interest earned on a 2 Year bond that’s renewed 15 times in the future).

A Unique Intervention

The US government buying back Treasuries is a relatively rare phenomenon. The most recent large buyback occurred in the 2000 to 2002 period, when $67.5bln of bonds were bought back. At the time, the outstanding debt was around $5.6trln, meaning around 1% of bonds were bought back. A recent paper by Connolly and Struby (2024) (“CS”) investigated that episode and found that the impact of the buyback was noticeable – yields in the bonds that were bought back fell by about 0.95%.

The current Treasury buyback policy, which started around 2023, is of a much lower scale. For it to be comparable to the 2000-2002 episode, around $400bln worth of bonds need to be bought (compared to the current $16bln per quarter buyback cap). CS, in their paper, actually discussed this specific buyback policy (prior to the announced increase) and found that the impact on yields will be zero. Which is basically what happened.

This new policy is unlikely to ‘fix’ what Scott Bessent thinks is the problem. Realistically, yields on long-dated bonds will only fall if the government enacts policies that reduce inflation. This includes ending the tariffs, easing immigration and ending the war. His government right now, however, appears to not be willing to do that. More concerningly, the government appears to not realize that these other factors are the problem.

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe to Our Newsletter

Share

Subscribe to Our Newsletter

Recently, the media outlets have stated that polls have again missed certain US elections with a special focus on the Democratic primaries in Michigan and Wisconsin. Media outlets are even questioning the reliability of polls. Given that elections are a popular topic, I wanted to clarify how polls should be interpreted and why the media headlines, as well post-election analyses, demonstrate a serious misunderstanding of what polls are.1

What Polls Actually Measure

The biggest confusion with polls is that polls are assumed to be direct reflections of what will happen on election day – if a poll shows a candidate ‘leading by 10%, then on election day the candidate should more or less win by such a margin’. This is incorrect – a poll is not a reflection or a forecast of what will happen on election day.

Polling Questions

A poll typically asks who would you vote for if the election were to happen today. Firstly, unlike an actual election, a poll is a low-stakes affair, and therefore people might not have fully considered what they would actually do on election day. Secondly, some people pick ‘undecided’ in polls. On election day, there is generally no such thing as an ‘undecided’ voter.2 Lastly, some people in polls pick non-leading candidates. On election day, they might change their vote to a leading candidate for strategic reasons.

Weighing the Sample

Once the pollster collects responses, the pollster then needs to appropriately ‘weigh’ the sample. That’s because the specific people that responded to a poll might not be representative of the group the pollster is trying to capture (for example, maybe too few young people responded in the sample).

But one question that arises – what group is the poll trying to capture – is it the general public or is it the people that will vote on election day? This is an important distinction, as it can change poll numbers significantly.

Poll Numbers

Once the pollster re-weighs the responses it got, the pollster publishes the polling numbers. Based on the steps above, the poll numbers are a snapshot of how the wider public (or voting public) feels about the election candidates at the time of the poll.

Election Day Forecasting

Instead of treating polls like a barometer, both the media and pollsters write headlines that view polls as if they are equivalent to what will happen on election day. For example:

The above language implies a forecast of what will happen on election day and that a candidate, El-Sayed, is ‘leading’ by 7 points. This headline is especially nonsensical, as a closer look at the poll shows that one candidate has 46%, while the other candidate has 39%.

So what about the 15% of undecideds – what will they do on election day? The headline above assumes that they will split perfectly evenly between the two candidates. Do you think that’s a logical assumption?

Forecasting

This is where forecasting comes in. Forecasting is modeling what exactly will happen on election day. That means predicting how many people will turnout, which people will turnout, and how each of them will vote. Polling can help inform our predictions, but it is NOT a prediction itself. A poll should be an input into a forecast model, and not the model itself.

But the media (and even pollsters) present polls as a forecast. There are plenty of elements that need to considered for a good election day forecast:

• Will everyone vote as they said in a poll? Will we see consolidation towards leading candidates?

• Who are the undecideds and will they show up to vote?

• Are any groups more likely to have a higher/lower turnout than typically assumed?

• Between the current poll and election day, what will candidates do (election campaigning, endorsements, new policy promises, etc.)?

Examples

If we take the above poll and headline as an example (let’s simplify by assuming 100 voters):

• Candidate A: 46% – (46 voters)

• Candidate B: 39% – (39 voters)

• Undecided: 15% – (15 voters)

Suppose Candidate A has a younger voter base, who are typically assumed less reliable to show up on election day. Let’s assume 3 of Candidate A’s voters do not go to the election. Next, let’s assume that, of the undecideds, 4 do not go to the election at all, while 8 vote for Candidate B and 3 vote for Candidate A. The election result will be as follows:

• Candidate A: 46 votes (49.5%)

• Candidate B: 47 votes (50.5%)

Candidate B wins the election. The media headlines will focus on “how did the poll miss this election”, as originally Candidate A was leading by 7 points, but saw an 8 point reversal.3

To belabor the point, let’s change one of our assumptions – suppose only 1 Candidate A voter does not go to the election. The result would then be:

• Candidate A: 48 votes (50.5%)

• Candidate B: 47 votes (49.5%)

Media headlines will again claim polling misses. But in both examples (and especially the second one), I hope it’s clear that the poll itself did nothing wrong. There was no polling mistake. The mistake was using the poll as a forecast!4

How I Use Polls

Hopefully the above clarifies how polling vs forecasting differs. Now, I doubt many of us have time to do deep dives and build our own forecasts. So here is how I quickly interpret polls for myself:

• If polls consistently show a candidate getting above 50%, then the candidate is almost guaranteed a win;

• If polls consistently show a candidate in the high 40s (47%+) and there are at least 6% undecided, the candidate has a very high chance of winning come election day;

• Anything below 47% (regardless of “lead”) is unknown and would require a deep dive to forecast the outcome.

I know the above guidelines are very crude and probably could be refined more, but that’s for a different writer to analyze. , whom I recommend on this topic, put together a historic table for US elections, which shows why ‘polling leads’ often do not mean much. The table below presents how often a candidate with a given polling support level (first column) and polling ‘lead’ actually won.

The conclusion one should draw from this is that ‘leading’ by 1% and having 49% support is better than leading by 7% and having 40% support – something that might seem counterintuitive.

In the appendix, I discuss two recent examples of alleged ‘polling misses’ and how I applied my rule of thumb.

Don’t Criticize Pollsters For Margin Misses

The point of this piece is to discourage criticism of polls by looking at polling margins. Polls are either conducted properly or improperly. They should not be judged against the actual results. These types of criticism also create incentives for pollsters to start targeting margin as the main metric rather than conducting a proper poll. And this may have actually happened recently, with a pollster adjusting results to more closely align with what they think will truly happen on election day:

TPSI [The Public Sentiment Institute] said in its ‘accuracy review’ of the survey that an error occurred after the firm applied an ‘experimental layer’ that applied previous survey results to the poll after its results ‘did not align with our read of the race’.

‘Acting on that conclusion, we identified respondents who selected [Florida Lt. Gov Jay] Collins but whose broader profile aligned with a competing [candidate], and leaned those respondents toward James Fishback’, the firm wrote in this document.

Just like pollsters, economists are also not in the forecasting business. We often get criticized, because our results ‘did not align with what happened’. As with polls, the real issue is misunderstanding what economists do and how to apply their work, not with the work itself.

Please let me know what you thought of the article length:

Appendix – Real Life Examples

Two recent elections restarted many of the polling-miss headlines – the senate democratic primary in Michigan and the governor democratic primary in Wisconsin.

Michigan

In Michigan, one candidate had a ‘polling lead’ of double digits, but in the end, the leading candidate won by 1 point. Here were the polls:

Using my own rule of thumb, as El-Sayed was polling in 50%+, I would say he was ‘guaranteed’ to win. Which he did. The fact that the result was ‘closer’ than the polls, simply suggests that undecideds split heavily in favor of Stevens.

Wisconsin

In Wisconsin, one candidate had an average ‘polling lead’ of over 15%. In the end, they lost by 0.5%. Here were the polls.

Polls had Hong in the high 30s/low 40s. Using my rule of thumb, I would not be able to say whether Hong would win or not without taking a deeper dive. Even if we use the naive5 historic trends, as in the table provided by Carl Allen above – Crowley had an ~1 in 4 chance of winning.

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe to Our Newsletter

Share

Subscribe to Our Newsletter

In the next few months, I will be going full time and turning on a paid feature. If you’d like to help me with some feedback (and a chance to win a 3-Month Paid Subscription tier) – please fill out the questionnaire below:

Nominal News Reader Questionnaire

Changing jobs can be both exciting and scary. As I am currently in that process, I have definitely already experienced both feelings. Financial security is always on one’s mind and I am fortunate to have some, as it allows me to explore going full time on Nominal News.

Governments provide the public such financial security through unemployment insurance (UI). It turns out also that UI can be an important factor driving economic growth as shown in a recent paper by Clymo, Denderski Mercan and Schoefer (2026) (“CDMS”).

Excessive Caution

An employed individual may naturally be worried about switching jobs. Going to a new job comes with inherent risks, one of which is what if the new job doesn’t work out and ends up terminating you. Faced with this uncertain risk of termination at a new job, a worker may be less inclined to switch jobs. This excessive, yet reasonable, caution can lead to worse outcomes for all.

UI may help with alleviating some of this risk. How?

Unemployment Insurance

To model the impact of UI, the authors, CDMS, assume a job is defined by two characteristics:

• how productive is it (for example, how much revenue it will generate)

• what is the likelihood the job will be terminated (e.g. how likely is the company to survive)

The wage a worker will receive is a function of these two things.1

The worker can decide whether to accept a job offer or not. If the worker accepts the offer, they receive a wage; if not, they receive UI.

Whilst on the job, the worker can receive job offers from other firms as well. As before, the job offer they get is described by the productivity level and job security level.

Intuition

It is intuitive that a worker can be indifferent between two jobs – one job that offers higher wages but lower job security is of equivalent value to a job that offers lower wages but higher job security.

Since the biggest worry for a person taking a higher risk job is becoming unemployed, UI can help mitigate this fear by providing a worker with some money in case they do become unemployed. The higher the UI, the more likely a worker should be willing to take a riskier job.

CDMS formally analyzed this with a survey2 to elicit how people view the wage vs security trade off. In this survey. CDMS found that:

• for a 1 percentage point higher probability of job loss, workers on average demand 1.63% higher wage;

• For an identical 1 percentage point higher probability of job loss, workers in higher paying jobs require a 1.9% wage increase compared to workers in lower paying jobs, who only require a 1.3% wage increase;

• For an identical 1 percentage point higher probability of job loss, workers in safer jobs require a 2.1% increase in wage increase compared to workers in riskier jobs who only require a 0.96% wage increase.

Consumption and Saving

Lastly, as is classic with most economic models, the worker chooses how much to consume each period and how much to save. Saving acts as a basic form of self-insurance from unemployment.

The CDMS model distills the world into a simple yet reasonable reflection of the world, as workers:

• Make a trade off between wages and job security;

• Consume or save.

Job Creation

The really interesting element of the CDMS model is that the decisions workers make also influence job creation. If workers are unlikely to switch jobs, firms are less likely to form and post job offers (posting a job entails a cost), as they are less likely to find a worker willing to join them.

At the same time, with higher unemployment insurance, workers are also less likely to accept jobs, again discouraging firms from posting jobs. This implies that there is an optimal level of UI, as both effects need to be balanced.

Bringing in Data

As the last step, CDMS calibrated the model to US data. In the US, unemployment insurance typically replaces around 40% of an unemployed person’s previous wage (this is called the replacement rate).

CDMS ran a counterfactual analysis by dropping unemployment insurance to about a 10% replacement rate. In this scenario, overall US productivity would fall by about 1.3%, while output would also fall by around 0.4%.

This would be primarily driven by the fact that job to job transitions collapse. The cost of becoming unemployed is simply too high. In turn, there are fewer risky-type firms entering the economy, which is why productivity and output fall.

Recessions

The really interesting situation arises during recessions. During recessions, the likelihood of becoming unemployed typically rises significantly for every type of job. This makes safer jobs even more desirable and reduces the willingness to switch jobs, as workers ‘flee to safety’. In turn, this rational, albeit conservative, behavior may further exacerbate the recession. Thus, CDMS propose a policy where the replacement rate of unemployment insurance is increased during recessions, to offset the effect of flight to safety.

For example, during the 2008 recession, to fully offset the flight to safety, the UI replacement rate of income would need to be increased temporarily by 50% to a 60% replacement rate.This higher replacement rate encourages workers to switch into riskier jobs, restoring productivity growth. It does result in a higher level of unemployment, since workers now are compensated better in unemployment, and thus, are more picky with accepting jobs.3

Unemployment Insurance – An Important Policy

Unemployment benefits are often criticized. Any expansions of UI, such as those during the Covid pandemic, are even more so criticized. The work by CDMS, however, shows how UI expansion in recessions might actually be an optimal policy. This is because people are generally risk averse, which can lead to sub-optimal, albeit correct, decision making. UI is needed to ‘correct’ for that risk aversion and encourage people to make decisions that benefit everyone.4

Moreover, CDMS show us that reducing UI benefits may actually reduce the number of firms willing to offer jobs. That’s because firms understand that employed workers are more fearful of switching jobs given a higher risk of unemployment.

The CDMS paper has also shown us how important it is to really consider people’s behavior. If we were to completely ignore the fact that people take into account the likelihood of getting laid off, our view of UI would be completely distorted and make the program look less beneficial.5

Please let me know what you thought of the article length:

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe to Our Newsletter

Share

Subscribe to Our Newsletter

In the next few months, I will be going full time and turning on a paid feature. If you’d like to help me with some feedback (and a chance to win a 3-Month Paid Subscription tier) – please fill out the questionnaire below:

Nominal News Reader Questionnaire

I recently had a conversation on Substack Notes about economists being bad at predictions. I agreed. But that’s because economists are not in the business of prediction, but something slightly different – causation.

Prediction vs Causation

New York City recently implemented congestion pricing (a toll to enter the city center). Prior to the launch of congestion pricing, 730,000 vehicles would enter the congestion zone daily. If I were to tell you that economics research tells us that the implementation of the toll would reduce the number of cars entering the zone by 10% – how would the number of vehicles entering the zone change one year after the implementation of the toll:

• A: 10% lower

• B: the same

• C: More than 10% lower

• D: Can’t tell

The correct answer, from this list, is D – can’t tell. How come?

A Subtle Nuance

Most economics questions can be boiled down to one formula1:

• Y is the outcome variable that we are interested to see how it changes. In our current example, it’s the number of cars entering the congestion zone.

• X is the policy variable that we can control. In our case, the toll.

• Z is everything else that also impacts the number of cars entering the zone. For example, this can be things like population changes, big events like the world cup/concerts, weather patterns, etc.

• Beta is ultimately what economists want to establish – what is the effect of changing policy X on Y. In our hypothetical example, I said that if we enact the $9 congestion toll, the number of cars entering the zone will fall by 10%. That would be the estimate of Beta.

So why can’t we say that the year after congestion pricing is enacted, the number of cars entering the congestion would go down by 10%? That’s because the things that are in Z can change next year, independently of our policy. Thus, the number of cars entering the zone can really take on any value depending on how Z changes.

However, if somehow, nothing in Z changed year to year – Z remained constant – then Y would fall by 10%. This is exactly what economists do say – the $9 toll would reduce congestion by 10%, holding all else constant (i.e. nothing in Z changes).

So what if we implement the congestion toll and next year we observe 600,000 vehicle entries? Since the toll only reduced vehicle entries by 10% of 730,000, the congestion toll only accounts for 73,000 of the 130,000 fall in vehicle entries. That means that something in Z must have also changed that pushed vehicle entries down.

Counterfactual

Another way to think about this is through a counterfactual approach. Had we not implemented the congestion toll, there would have been 73,000 extra vehicle entries. So instead of 600,000 that we actually observed, we would have had 673,000 vehicle entries.

Prediction

People are generally interested in prediction rather than causality. In our case, the news headlines will always simply report the total vehicle entries and that’s what people will focus on. From this number, everyone will erroneously make judgments.

In our little formula above, it means the news and people are really interested in variable Y, and not Beta (the number economists care about).

To predict Y accurately, you not only need to know Beta and X, but also take a stab at forecasting what will happen with all the Z variables. Economists typically do not do that and our tools are generally not good for this. If you want to get predictions of Y, do not ask economists. We can only tell you how changing X will impact Y (the Beta). Our tools are designed to estimate this Beta number well.

Causality

I hope the above explains the nuance between causality and prediction. Causality can inform predictions, but it is not a prediction itself.

I like to think of causality as fundamentally similar to the rules of physics. The causal impact estimated by economists in any context is no different from, say, Newton’s laws of gravity. I do not view the law of gravity as a form of prediction.

The only reason why economists hesitate to call anything ‘laws’ in our field is because establishing laws in economics is much harder than in physics. That’s because we cannot control many things like physicists can – if they want to run an experiment, physicists can often build a physical tool that removes all or most of Z (the large hadron collider in Europe is a prime example of such an item).

Since in economics, we study people, such experiments cannot be run.

Human Nature

The problem economists face is that the economics’ way of thinking is not intuitive. People always want to know what Y will be and not how X impacts Y (Beta). One of the first questions economists get in social settings is always of the type “where do you think the economy is going?”2. I cannot answer that question. I can, however, tell you that if you enact congestion pricing you will have 10% fewer cars, holding all else constant.

People want to know what GDP will be next year, what inflation will be, what their income will be. They directly attribute what they end up observing to the policies that were enacted.

But this is incorrect. Outcomes are impacted by policy, but the specific level of the outcome (actual GDP, inflation, income, etc.) depends on a lot more than just one policy. Often, these non-policies dominate the impact of any policy.3 It still makes the policy worthwhile, but we shouldn’t expect massive changes when things are influenced from a thousand different sources.

P.S. Economists and Predictions

One thing I wanted to add after finishing this piece, is that lately there has been a noticeable increase in economists making predictions, especially around genAI. Recently, 200 economists signed a letter warning about the dangers of AI. It is also not uncommon to find futuristic predictions from economists in social media. I don’t have issues with people discussing the future, but these predictions are personal opinions, rather than causal statements backed by economics research. But, as was the case with the letter, there is significant emphasis on the fact that economists are the ones making these predictions, with the purpose of giving these statements additional validity. This implicitly presumes that economists are better futurists than any other profession, which we are not.

As I laid out in this piece, there’s nothing about economics that really makes us any better at prediction.

Please let me know what you thought of the article length:

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe to Our Newsletter

Share

Subscribe to Our Newsletter

Summary

• The FOMC (the interest rate setting committee) decided by a split vote (9-3 ) to keep interest rates between 3.5% and 3.75%.

• The 3 voters wanted to increase interest rates to 3.75% - 4.00%.

• Warsh’s press conference after the decision had a lot of contradictions and somewhat concerning statements.

Commentary

The Decision

In advance of today’s meeting, the probability of rate hike was quite high (between 30% and 40%), which made this meeting more important than normal.

The decision by the FOMC to keep rates steady was widely expected, although the fact that 3 voting members dissented is something we do not see often. The reason for the dissent was most likely driven by the fact that forecasts for June inflation remain elevated.

Current forecasts suggest that 12‑month inflation will be at 3.3% for core (excluding food and energy) PCE (Personal Consumption Expenditure) and 3.7% for headline PCE. On a month-to-month basis, Core PCE inflation is expected to be up in June by 0.18%, which gives an annualized inflation rate of around 2.2%. Note, that although the monthly rate is in-line with the the 2% target, recent US government decisions around the war in Iran and tariff re-enactment suggest future inflation pressures.

Last time, when the FOMC met in June, I mentioned that it was surprising that the vote to hold interest rates steady was unanimous (12-0). This time, no such surprise. I could believe that the vote could have been even closer given some recent statements made by FOMC voters.

What’s interesting is that Kevin Warsh did not vote for the hike. Although his individual vote to hold rates steady was expected, there is a bit of tension between his rhetoric of committing to a 2% inflation target, which would probably necessitate an interest rate increase, and his prior statements that interest rates need to be cut.

For example, from a statement he made 1 year ago.

“Their hesitancy to cut rates, I think, is actually ... quite a mark against them,” Warsh told CNBC. “The specter of the miss they made on inflation, it has stuck with them. So one of the reasons why the president, I think, is right to be pushing the Fed publicly is we need regime change in the conduct of policy.”

This tension between his prior statements and current economic reality may be his influencing voting behavior.

The Statement

Below is the latest FOMC statement, with changes highlighted in comparison to the June FOMC statement:

The statement has only minor changes referencing the fact that there were dissenters.

Although last time I interpreted the language around productivity growth and capital investment as suggesting what was discussed specifically at the meeting, my current reading is that it is a boilerplate statement. Therefore, not much can be gleamed from this statement.

Kevin Warsh Press Conference

Kevin Warsh had held his press conference after the decision. It was quite astounding - and not in a good way.

Prepared Statement (57:20 - Video)

The beginning of the prepared statement closely followed the officially published statement. Kevin Warsh reiterated that he will not provide forward guidance, because it would not be prudent during “these uncertain times” (note - this contradicts something he said later).

Kevin Warsh then bizarrely focused on criticizing the work done by the Federal Reserve over the last 5 years. Some quotes:

• “5 years of high inflation has left a mistaken impression that’s hard to shake'“

• “we have begun a new chapter and we understand that the 5 plus years of inflation above target cannot be cured in 9 weeks”

First, many of the current FOMC members were also FOMC members during these 5 years. Thus, Warsh appears to be insinuating that they’ve done a bad job.

Moreover, Warsh completely ignores the fact that when controlling for lagged rent data (something Warsh would advocate for with his emphasis on real-time data), inflation did drop to 2%. Thus, the criticism can be considered unfair.

The criticisms did feel more political in nature, rather than economic.

On Forward Guidance

From the press conference, it is clear that Warsh is laser focused on one thing - removing ‘forward guidance’.

As a quick reminder, forward guidance is the practice where central banks telegraph future monetary policy decisions. The purpose of forward guidance is increase central bank credibility in front of market actors. With this credibility, central banks do not have to make drastic policy changes to get inflation under control. That’s because market participants will anticipate what the central bank will do.

Warsh believes markets should not be informed about the exact choices the Federal Reserve will make. That’s because he believes it should be the other way around - market decisions should be act as an information signal for the Federal Reserve. Forward guidance influences market decisions and, thus, contaminates this signal that markets send to the Federal Reserve.

The problem I have with this idea is that central bank credibility is hard to establish, while the market signal can probably be extracted by other ways (e.g. directly talking to businesses). Moreover, I’m not sure why markets reacting pre-emptively is inherently a bad thing.

Separately, in a latter part of the conference, Warsh recounts how during a crisis (referencing the 2008 financial crisis), forward guidance was needed to help ease concerns. But his belief is that during ‘benign’ periods, forward guidance is bad.

This seems contradictory to me, especially since he has called the current times as ‘uncertain’.

Market Reaction - Another Contradiction (1:05:20 - Video)

Warsh puts a lot of emphasis on not having forward guidance because he wants unfiltered market data (i.e. not contaminated by the Fed’s future decisions).

A reporter then asked about how he interprets the financial market data which currently appears to suggest that interest rates should be higher. Warsh responds:

“Interpreting markets is an imperfect business”

And does not provide any additional relevant response to the question.

This wasn’t a promising answer. Quite the contrary - it’s a bit concerning.

If he wants financial market data, but he doesn’t know how to interpret it, then I’m not sure what’s the point…

Data Dependence - Another Contradiction (1:08:45 - Video)

Warsh was also confusing about using data. From the previous conference, Warsh emphasized the need for new data. During this press conference, Warsh says he does not like “data dependence”, but he cares about “trends on the data”.

Again, I’m stumped. What is the difference between data dependence and the trends in the data? Economists always look at trends and not just indvidual data points, if that is what he’s referencing.

The concerning interpretation of this statement is that Warsh is likely to pick and choose what data fits a particular narrative, he may want.

Comments on the FOMC Members

Lastly, throughout the press conference, Warsh kept reiterating how he has become more confident in the group working with him at the FOMC. He feels more certain than ever that the FOMC group will get inflation to the 2% rate.

This sounds like a harmless statement, but I found it a bit peculiar.

Many of the FOMC members have been on the FOMC for some time. By emphasizing that he feels more confident in the group, Warsh makes it sound like he was skeptical when he joined the Federal Reserve that these people would be good professionals. The repeated emphasis that he now trusts them after 9 weeks on the jobs seemed peculiar to me.

An Even More Concerning Press Conference Than in June

The decision to keep rates steady was not a big surprise. Neither was the fact that there were dissenters. The press conference, however, was something much more surprising.

It appears that Warsh is concerned by inflation, but does not want to raise interest rates, believing that just higher US treasury rates will suffice in slowing down the economy (which in turn, would reduce inflation). If that were the case, then we wouldn’t even need the Federal Reserve, as markets could control inflation by themselves. But we know that this is not true.

More broadly, although mistakes in monetary policy today might not have grave consequences, my concerns are what will Kevin Warsh do during a more pronounced crisis - such as the Great Recession of 2008 or the Covid pandemic.

His answers to questions left a lot to be desired, and it was quite clear that there were lots of contradictions in these answers. This does not bode well for crisis situations.

Perhaps confusion was the intent - if what the Fed says is contradictory, then we can’t have ‘forward guidance’.

Finally, let me just leave you with the reaction of the US 10 Year Treasury yield at the time of writing this article:

I think the signal here is that the markets really didn’t like this press conference.

This has been a special edition Nominal News article covering a very recent development. Please let me know your thoughts on this feature, and what would you like to see covered, in the comment section below.

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe to Our Newsletter

Share

Subscribe to Our Newsletter

In the next few months, I will be going full time and turning on a paid feature. If you’d like to help me with some feedback (and a chance to win a 3-Month Paid Subscription tier) – please fill out the questionnaire below:

Nominal News Reader Questionnaire

Tariffs continue to dominate US economic headlines, especially as the US has recently reinstated them. This entire tariff episode, which started in 2017, has reinvigorated the economics debate around ‘free trade’. All of a sudden, even economists that would previously unequivocally support ‘free trade’, are starting to suggest that tariffs may have a place as a polity tool. One recent paper by Werning and Costinot (2026) (“WC”) goes over the theoretical instances when tariffs may be beneficial.

The Problems with the Term “Free Trade”

First – I’m glad that ‘free trade’ is being revisited. In 2023, I raised the issue of ‘free trade’ being a misnomer, as any trade agreement is more than just tariffs or quotas. From our piece:

“The concept of ‘free trade’ is too vague to be meaningful. Trade agreements involve a lot of negotiation on issues that are beyond what we used to consider as ‘free trade’ – removing tariffs and quotas. The other components, especially regulatory issues, are just as important, not only for the negotiating countries, but even for the wider environment. But these tools are also manipulated by countries and special interest groups to create benefits for themselves. Economists should no longer be quick to say ‘free trade’ is good.”

So I am glad to see that the economic discussion is now focusing on whether maximizing trade improves welfare, and WC, in their paper, highlight what to consider before we can determine that.

Setting the Stage

To start off, it’s important to set the definition of a tariff:

When an activity is taxed, the activity (assuming everything else does not change) is reduced. Thus, tariffs reduce the amount of imports that occur. Many criticisms of tariffs focus on this – we should not be dissuading ‘trade’, as more ‘trade’ is typically welfare improving.1 So when should we dissuade trade?

Situations of Social Harm

WC start off by looking at cases where trade may cause some social harm. This social harm is something not taken into account by the individuals entering into a trade arrangement. A classic example of this is pollution. If I can import a good cheaper, I will do so, even if its production generates more pollution than if I had bought it domestically. I do not take into account the social harm of pollution, which is a cost on all members of society.

This means that absent a tariff (tax) on the import of this good, I will be consuming too much of this imported good from a society perspective given how much pollution I generate. A tariff would reduce my consumption of the good, which would in turn reduce the amount of pollution generated.

There are many examples of things we do not take into account when entering any trade transaction:

• Pollution;

• Personal moral standards – for example, if the good utilized child labor in its production;

• Geopolitical considerations – a topic that has recently become popular in tariff contexts with regards to ‘national security’;

• Reduced domestic knowledge – if in an industry, workers learn by doing (like semiconductor manufacturing), trade may reduce domestic production, which decreases learning on the job.

Tariffs, in theory, can mitigate some of these ‘social harms’ from trade. Before we evaluate the impact of tariffs, let’s turn to another case where tariffs may play a role.

Domestic and International Redistribution

Another example of when tariffs may be beneficial, mentioned by WC, is if they are intended to redistribute income. This income can be redistributed in three ways:

• From consumers to government – similar to a standard tax;

• From one group of consumers to one group of producers/workers;

• From one country to another.

The first case is straightforward – a tariff raises revenues for the government by collecting money from consumers.

To understand the second case, let’s assume the government puts tariffs on textile products. This ‘protects’ domestic textile workers. By increasing the price of imported textiles, domestic textiles, which were previously more expensive, may now be relatively cheaper. This in turn increases demand for domestic textiles, which increases wages for textile workers (and profits for textile manufacturing owners). Thus, domestic consumers end up paying more for the textiles, but the textile industry benefits.

The final case can be thought of commonly with the ‘foreigners will pay for the tariffs’ argument (a common argument used by tariff proponents). Basically, by imposing a tariff, a foreign exporter may need to reduce the price they offer their good to the importers. In this way, ‘foreigners’ transfer money to the domestic country, as they reduce the price. It is worth noting that recent research on current tariffs has shown very little to no response to price by foreign exporters.

So Can Tariffs Be Good? A Rebuttal

The above cases mentioned by WC give a good overview of situations of when tariffs may be welfare enhancing. But let’s see if tariffs are actually a solution in these instances.

Social Harm Cases

If we look at the first instance of social harm (e.g. pollution), tariffs can reduce social harms. BUT, this is only true if a similar tax is levied on domestic goods. For example, if domestic production is more polluting than foreign production, then a tariff will actually make the social harm worse.

The key element to note is imports and domestic production are never truly the same. Even if the product itself is the same, for example, steel, if the domestic or imported version was produced with more pollution, can they really be considered the same? Basically, are low pollution steel and high pollution steel interchangeable or should we differentiate them? Ensuring taxes and regulations are uniform on all goods is a complex issue, and are often the reason trade negotiations fail. But without this standardization of goods, trade agreements may lead to harmful outcomes, as was the case with San Diego fisheries.2

So just tariffs are not a good solution – a far better solution is a standardized tax and regulations to make goods truly similar.

Learning By Doing

In the case where there is a learning benefit from producing a good domestically, tariffs are probably a suboptimal solution. Instead, providing subsidies to the domestic industry would likely be superior. That’s because tariffs protect domestic producers from actual competition, which discourages domestic innovation.

We can think of tariffs as adding a weighted vest to your running competitor – it slows them down, but doesn’t improve you. On the other hand, subsidies, especially ones with an expiry date, encourage domestic producers to get better at production, and compete with foreign producers.

Therefore, although tariffs can help with developing nascent domestic industries, other solutions are likely to be superior.

Domestic Re-Distribution

Regarding domestic re-distribution, tariffs can help re-distribute income domestically. But as WC say in their paper, citing Dani Rodrik (another economist):

“Saying that trade policy exists because it serves to transfer income to favored groups is a bit like saying Sir Edmund Hillary climbed Mt. Everest because he wanted to get some mountain air. There was surely an easier way of accomplishing that objective.”

However, as WC point out, even though tariffs are a bad tool for this type of re-distribution, there might not be any other politically feasible way. Opening up to trade creates winners and losers domestically. In a perfect world, we would re-distribute the trade gains from the winners to the losers, but often, this does not occur.

In one of my research papers, I worked on the idea of gradual trade liberalization (for example, by slowly lifting a quota over several decades) and showed that it may help protect workers in an industry until they retire. Through this gradual liberalization, you effectively reduce the gains of winners and transfer them to the ‘losers’ from trade.

Trade Protectionism

The WC research is a very helpful theoretical guide, but I view the research piece as a double-edged sword. It has its benefits, as the paper acts as a “theoretical compass” to evaluate when a government may need to tax or subsidize trade. As mentioned previously, I have never been a fan of ‘free trade’, as the term hides a lot of nuance around regulations, rights and real costs, such as pollution. ‘Free trade’ oversimplifies what actual trade is.

At the same time, however, some ideas in this paper may be ripe for abuse. WC quote economist, Francis Edgeworth, commented on researching the theoretical benefits of tariffs:

“The direct [benefit]... of the theory is likely to be small. But it is to be feared that its abuse will be considerable . . . Let us admire the skill of the analyst, but label the subject of his investigation POISON.”

Tariffs are almost always a suboptimal policy – solely placing a tax on imports is costly and welfare-reducing compared to alternative policies (subsidies, re-distribution). This is especially true in a globalized world, where no importing or exporting country is sufficiently large to impact the market.

In my opinion, a government choosing to implement tariffs, even in their best use case, already signals policy defeat, as the government failed to convince the public of the benefits of superior policies. Rather than focusing their energies on inherently poor policies, governments and politicians should push for far superior solutions, of which economists have plenty.

If you would like to support us in reaching our subscriber goal of 7,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!

Subscribe to Our Newsletter

Share