Desktop wide screen and notebook header

  • Register and access free online courses on Macroeconomics, Microeconomics, Introduction to Valuation, Introduction to Evaluation

Tablets header

  • Register and access free online courses on Macroeconomics, Microeconomics, Introduction to Valuation, Introduction to Evaluation

Mobile header

  • Register and access free online courses on Macroeconomics, Microeconomics, Introduction to Valuation, Introduction to Evaluation

Productivity, ChatGPT and Steady State

This article discusses the recent turmoil in the US financial markets, the Fed’s response to inflation, the Production Function, and the Solow Model. This piece uses ChatGPT to explain the concept of Steady State. Additionally, it takes an interesting look at both the pros and cons of artificial intelligence, with a peek at behavioral finance guru Hyman Minsky.

The recent turmoil in the financial markets, caused by developments at Silicon Valley Bank, Signature Bank, and Credit Suisse, highlights risk management and compliance failures. These failures also showcase the inherent fragility within the banking sector, as epitomized by Dr. Hyman Minsky. In his landmark paper, “A Theory of Systemic Fragility,” Minsky proposed hedge financing, speculative financing, and Ponzi schemes. As banks are more akin to speculative financing, Minsky’s recommendation to look at liability structures and asset equity ratios, especially of banks, is of utmost priority. The Fed’s actions have calmed the markets and assured that the US banking system is robust. However, repeated banking crises have a high global negative externality that reinforces the fragility of the banking system. JPMorgan reported better-than-expected results, which helps the Fed’s case. Still, the asset management mismatch and the exponential effect of fractional banking reinforce the need for tighter regulation among banks.

The US inflation rate reduced to 5 percent last month. The minutes of the Federal Open Market Committee on March 21-22 discussed poor liquidity and high implied volatility due to the closure of Silicon Valley Bank and Signature Bank, but the banking crisis in the US is far from a systemic risk. According to the Bureau of Labor Statistics, the Producer Price Index fell 0.5 percent in March, seasonally adjusted. Two-thirds of the decrease in demand is due to a 1 percent decrease in prices for final demand goods, suggesting that there would be a downward movement in US inflation going forward. In March, core inflation increased by 0.4 percent, up 5.6 percent over the year, while the CPI for all urban consumers increased 0.1 percent (seasonally adjusted), rising by 5 percent over the last 12 months. Inflation in the US is showing a downward trend, which is positive for the US. The data also showed that the PCE (i.e., consumer price inflation) somewhat slowed in February. According to FOMC minutes, the gains in consumer spending in recent months are offset by the slowdown in business fixed investment and residential investment. The increase in the borrowing rate (effective fund rate at 4.83 percent as of May 1, 2023), complemented by the reduction of borrowing by banks, will further enhance the slowdown in business and residential investment. The minutes clearly note that the reduction in wage growth would translate into reducing the overall aggregate demand, lessening inflation as we move ahead. The Fed is confident that by next year, inflation would come down to 2 percent.

Consumer confidence in the US remains upbeat. The University of Michigan survey of consumers posted a 2.4 percent change month on month. The rise in year-end inflation expectations in April is due to the prevailing uncertainty in the economy. Consumer expectations are in line with the Fed’s view that inflation will be high in the short run but lower in the long run. Consumers are upbeat about the economy as current economic conditions came out stronger in April compared to last month. The unemployment rate in the US dropped to 3.5 percent on the back of a tight labor market. However, the effect of the banking crisis remains uncertain, and the FED believes that the unemployment rate could rise to 4.5 percent by this year’s end.  There will be tradeoff between Inflation and Unemployment Rate. A slowdown in wage growth and an increase in the unemployment rate could lower aggregate demand curve. In the short run, shifts in aggregate demand impact employment and output but not prices. In the long run, shifts in aggregate demand affect the price level. In the short run, prices are sticky, but in the long run, the aggregate supply curve is fixed by factors of production and the level of technology. With high inflation, participants at the FOMC meeting believe that a period of below-trend real GDP growth will lead to a better balancing of aggregate demand with aggregate supply, thereby reducing inflationary pressure.

Subscribe to the Applied Learning course to learn through various topics on Inflation. Check out the lesson on Phillips Curve.

Applied Learning

The middle road does not see the Fed reducing interest rates anytime soon. Inflation is high but showing a downward trend. The unemployment rate in the US is expected to increase, leading to a downward shift in aggregate demand. The non-accelerating rate of unemployment, also known as NAIRU or the natural rate of unemployment, is the lowest sustainable rate of unemployment without triggering inflation. As the unemployment rate rises, the aggregate demand of the economy goes down due to a reduction in consumption and wages. 



Note: This Natural Rate of Unemployment (Short Term) series is discontinued. Chart Source: U.S. Congressional Budget Office, Natural Rate of Unemployment (Short-Term) (DISCONTINUED) [NROUST], retrieved from FRED, Federal Reserve Bank of St. Louis;, May 2, 2023. Units:  Percent, Not Seasonally Adjusted

According to neoclassical economics, in the long run, output is fixed, and a downward shift in the demand curve only reduces the price level of the economy. This process will reduce the inflation rate, which is exactly what the Fed envisions. Monetary tightening would have little effect on the supply side of inflation due to the uncertainty of the longevity of the Russian-Ukraine conflict and the rising tensions between China and Taiwan.

The effect of the increase in policy rates is not easy to determine due to the lag time between the policy action and its intended consequences within an economy. Real GDP, inflation, and the unemployment rate are lag indicators. Another lag indicator is the Case Shiller Home Price Indexes. Developed by Allan Weiss, Karl Case, and Robert Shiller, the index uses repeat sales methodology. The S&P CoreLogic Case-Shiller 20-City Composite Home Price NSA Index seeks to measure the value of residential real estate in 20 major U.S. metropolitan areas: Atlanta, Boston, Charlotte, Chicago, Cleveland, Dallas, Denver, Detroit, Las Vegas, Los Angeles, Miami, Minneapolis, New York, Phoenix, Portland, San Diego, San Francisco, Seattle, Tampa, and Washington, D.C. This is discussed in the online macroeconomic course “Introducing Economic Fluctuations & Their Impact on the Economy.” A dip in the Case Shiller Home Price Index is a harbinger of a recession. The Fed is forecasting a mild recession that runs through 2025, and a fall in the Case Shiller index is pointing to an ongoing recession. Feel free to register for free and access the online course.

Chart Source: S&P Dow Jones Indices LLC, S&P/Case-Shiller U.S. National Home Price Index [CSUSHPINSA], retrieved from FRED, Federal Reserve Bank of St. Louis;, May 1, 2023. Index Jan 2000=100, Not Seasonally Adjusted, monthly.

Introducing Economic Fluctuations & Their Impact on Economy

Dr Robert Solow, ChatGPT and Rise in Productivity

Dr. Robert Solow, the father of economic growth theory and known for his contribution to economic theory, is the creator of the Solow-Swan growth model. This model explains the long-run economic growth of a country as a function of labor and capital, with the level of technology considered an exogenous factor. If you recall the production function, which is covered in The Middle Road’s online macroeconomics course, the level of technology, also known as total factor productivity, is exogenous, meaning it is not defined by the model. Swan worked independently to make contributions to the model.


The Solow Model explains how the factors of labor, capital, and savings rate, through the investment function and population growth, determine capital accumulation leading to economic growth in economies over the long run. Advanced economies eventually reach a steady state when increases in the capital-labor ratio plateau, while emerging economies experience high growth due to a rise in the capital-to-labor ratio, keeping the level of technology constant. The level of technology, however, is considered a constant outside of the model. The purpose of discussing the Solow Model is to introduce the concept of steady state. Topics such as convergence of economies will not be discussed.

Productivity of Labor per hour 2013=100. Chart The middle Road. Data Source: Bank of England, a millennium of  macroeconomic data. The graph was part of the lesson on Production Function from The middle Road. The revised version will be published.  

Empirical studies that have examined the contribution of TFP growth to economic growth have generally found that technological progress has been a significant contributor to economic growth over time. For example, a seminal study by Robert Solow in the 1950s found that TFP growth accounted for more than half of the growth in output per worker in the United States between 1909 and 1949.


More recent research has continued to highlight the important role of technology in driving economic growth. For example, a study by Dale Jorgenson and his colleagues estimated that TFP growth accounted for around 70% of the growth in output per worker in the United States between 1947 and 2007.


According to McKinsey’s research, TFP growth has been a significant contributor to global economic growth over the past several decades. The firm has estimated that TFP growth accounted for around 30% of the increase in global GDP between 1990 and 2014. McKinsey has also identified a number of factors that are driving TFP growth, including advances in digital technology and the increased availability of data and analytics. Refer to the Factors of Production lesson under Applied Learning where few more factors are included example data.  Source ChatGPT

Lets look at Aggregate Production Function first. 

Cobb Douglas Production Function

AKαL1-α  where K, L are fixed and exogenous. K : Structure, Equipment, Plant, Machinery etc.  L : Labor, A= Total Factor Productivity or Level of Technology.

Cobb Douglas Production Function measures the effect of productivity on the economy and the share of labor and capital has remained constant over the year i.e., 70% labor participation and 30% capital participation. This data is valid for US.

Chart Source: University of Groningen and University of California, Davis, Total Factor Productivity at Constant National Prices for United States [RTFPNAUSA632NRUG], retrieved from FRED, Federal Reserve Bank of St. Louis;, May 2, 2023. University of Groningen and University of California, Davis, Total Factor Productivity at Constant National Prices for Japan [RTFPNAJPA632NRUG], retrieved from FRED, Federal Reserve Bank of St. Louis;, May 2, 2023. Units:  Index 2017=1, Not Seasonally Adjusted, Frequency:  Annual

National Income = Real Labor Income + Real Capital Income

National Income is divided between payments to capital and labor. The size of payments depends on marginal products of labor and capital.  The share of labor and capital income remains the same regardless of level of national income.

Cobb Douglas Production Function has the following characteristics.  

  • Constant Returns to Scale
  • Diminishing Marginal Product of Labor
  • The marginal product of labor (MPL) indicates how much output increases for each additional unit of labor, holding other inputs constant
  • The slope of the production function, ΔYK as the labor increases, the marginal product of labor declines



Chart Source: @2015 Pearson Education, Inc All Rights Reserved Mishkin 

National Saving Investment Consumption Function

In the Solow Model below, the capital-worker ratio is derived by dividing output by labor to get the amount of capital per worker. All measurements are recorded per worker/labor. The output measured is output per worker. It is important to remember that in a closed economy, investment is equal to savings and output is the sum of capital and investment. Refer to the online macroeconomic course to know more. Therefore, investment is a function of the savings ratio of output, depicted as the investment function = sA1Kt0.3  where Kis the capital labor or capital worker ratio.

Image: The middle Road | Change in steady state due to technological advances 

At steady state investment per worker is equal to depreciation plus capital dilution since change in capital per worker is zero.

Depreciation plus Capital Dilution (at K* and K1*) = sA1Kt^0.3

As shown in the image, in the initial years, the capital-labor/worker ratio increases, which is measured through the slope of the investment function. When the economy approaches steady state at K1*, the slope of the investment function is zero, indicating that the increase in capital-worker ratio is also zero. We look at net investment, i.e., the investment function less depreciation and capital dilution. According to the Solow Model, an increase in the labor force lessens the capital per worker, whereas the Romer Model looks at the rise in population growth positively. Dr. Paul Romer, a Nobel laureate and presently a professor at New York University (also taught at the University of California Berkeley), proposed the Endogenous Growth Model, where the level of technology is explained by the model or system. These endogenous growth models differ from neoclassical growth models in that they explain economic growth as an endogenous outcome of an economy making technology and innovation part of the system. * Romer posits that technology, taken as a set of ideas, is non-rivalrous and non-physical, and along with knowledge spill-overs augments the level of technological progress within an economy over the long run.

Check out the Cart Here.

While the Solow Model suggests that high population growth (higher labor force) lowers the capital-worker ratio, keeping capital constant, and reducing output per worker, the endogenous growth model, on the other hand, looks at an increase in population growth as an enabler for total factor productivity through well-thought policies, for example, fostering growth in human capital through quality education and entrepreneurship. These policies will amplify productivity through innovation, generating an increase in returns model. The Endogenous Growth Model is a significant economic long-run model for policymakers to enable policies that promote technological advancement to increase productivity. For example, providing quality K12 and higher education, increasing research and development by the government, or promoting intellectual property rights through copyrights or patents to promote excludability of products, and facilitate an increase in investments in technology. Many of the innovations globally are funded by the government or government-led initiatives or agencies, for example, the U.S Defense-funded ARPANET that led to the invention of the internet or digitalization of payment services in selected countries.

The Origins of Endogenous Growth by Paul Romer is an excellent paper tracing the assumptions behind the endogenous growth model.


ChatGPT and Steady State  

ChatGPT is an excellent example of the rise in productivity within companies, keeping other factors constant. ChatGPT has revolutionized the automation of routine tasks, especially content creation across various themes, providing colossal value add for research, optimizing time, and aiding decision-making. For example, when the words “McKinsey on the contribution of total factor productivity on global growth” are fed to ChatGPT, it leads to concise output, a part of which is published above. The same words on Google would lead to a link to the McKinsey report. Another example is proofreading. The Middle Road used ChatGPT for proofreading material of this article, and the result is very impressive. ChatGPT is a technological leap akin to what Google did for the search and advertising industry.


Furthermore, ChatGPT’s success is propelling major technology companies to replicate its success by developing their own artificial intelligence systems that will become game-changers in driving employee productivity, leading to an increase in the capital-worker ratio and output per worker. ChatGPT is a major technological advancement shock. The use of ChatGPT is an excellent example of the rise in productivity within companies increasing the capital worker ratio and output per worker. The areas that ChatGPT will redefine in the long run are hard to comprehend, but the Middle Road envisions that ChatGPT will shift the investment curve up in the Solow Model discussed above. ChatGPT will shift the curve upwards, leading to a new steady state for the US economy, raising the capital-labor ratio. The free version of ChatGPT does not track real-time data; nonetheless, the product is a huge leap compared to its peers.

Back to Inflation

There is clearly a downward trend in inflation. Oil prices are lower on the back of a global slowdown, with Brent Oil Futures trading at less than $80 for the end of May settlement. The onset of a recession is clearly on the horizon; global macro risk and a slowdown of credit within the economy suggest a prevailing downward trend in inflation. The closure of two banks in the US shows the speculative financing substructure of banks. As interest rates rise, the margin of safety for banks reduces, with the excess of the present value of assets over liability diminishing or turning negative. An increase in interest rates increases the discounting factor, reducing the present value of cash flows. The Fed has been aggressive in using monetary policy as a tool for addressing inflation, coming out strong as a lender of last resort, but the lack of regulatory foresight in evaluating risk management systems of small banks needs to be addressed. Bank runs, such as the one at Silicon Valley Bank, follow herd behavior, a phenomenon regularly witnessed over time.

Pitfalls Ahead with AI

The effect of ChatGPT on productivity is expected to boost the GDP of the US economy. Technological advances, especially in the field of artificial intelligence, also have their own concerns. Geoffrey Hinton, also known as the ‘Godfather of Artificial Intelligence,‘ recently resigned from Google. Dr. Hinton, along with a couple of students at the University of Toronto, pioneered the use of artificial intelligence before moving to Google. Hinton’s resignation citing the dangers posed by AI is a significant red flag for policymakers to seriously consider the detrimental effects that AI may pose to humanity. ChatGPT’s big splash should be an eye-opener to judiciously manage technology upheaval for the better as we progress ahead.


Recommended Read

  • Macroeconomics Policy & Practice by Federic Mishkin

* one key significant difference.


Note: The Solow Model was corrected for the subscript error of the investment function. A revised chart of Total Factor Productivity is added. The steady model expanded for more clarity since its publication. 


%d bloggers like this: