Using Machine Learning and Big Data to Understand Micro Markets Vernon H. Budinger, CFA November 3, 2023

Overview

This paper focuses on a data/AI toolkit that marketing managers can use to understand the demographics in their market. While many AI projects rely on enormous data sets and new intimidating neural algorithms, Machine Learning provides informative, detailed assessments of consumer markets in economic micro-regions using large pools of free data and free software.  Such “humble” AI efforts can dramatically improve the reach and efficiency of marketing campaigns.

Targeted Market Segments.

There are many avenues for creating region-specific content and delivering that message to a specific census tract.

·        Facebook, Instagram, Twitter, and LinkedIn allow you to target a small region with posts.

·        Direct mail companies can create mail campaigns by census tract

·        Advertisements can be placed in community calendars and local news media.

·        Billboards reach specific market segments

The key is to segment your audience based on demographics, interests, and needs.  The Census Bureau now provides APIs (Application Programming Interface) to pull Census Data by tract or census block (smaller than a tract).  In addition, the American Community Survey is a U.S. Census service that updates the regional information yearly.  This data is free and is tagged with geolocation information.

The key is to segment your audience based on demographics, interests, and needs.  The Census Bureau now provides APIs (Application Programming Interface) to pull Census Data by tract or census block (smaller than a tract).  In addition, the American Community Survey is a U.S. Census service that updates the regional information yearly.  This data is free and is tagged with geolocation information.

This paper demonstrates how the U.S. Census/ACS study can used with freeware statistical/graphing packages to explore the ACS data and develop profiles of the consumers in the area.  The paper will show statistical/Machine Learning analysis of the data that provides deep insights into the demographic characteristics of each census tract.

Using Facebook to send content to target markets

Section II Market Analysis of North Bay California Counties

These tools were chosen because the data and analytics are accessible to small and medium-sized companies. They offer a better understanding of geo-demographic trends, improve customer experience, and build stronger relationships with the client base.  The analysis produces detailed quantitative measures of economic and demographic status as well as consumer behavior for micro-regions.

This analysis was adapted from Chapter 8 of Kyle Walker’s book “Analyzing U.S. Census Data.” This study focuses on median home value from the U.S. Census Bureau’s 5-year American Community Survey (5-year ACS).  While this data depends on estimates, it is more current than the Decennial Survey and has more data than the 1-year ACS, which only covers areas of population equal to or greater than 65,000. The more current 1-Year ACS would not cover any of the cities except Santa Rosa in Sonoma. 

Five County Demographic Comparison

Sonoma, Lake, Marin, Mendocino, & Napa Counties 

The data table for the five counties illustrates the challenges that marketing operations face in this complex region. Sonoma and Marin Counties in the southern part of the North Bay Region are wealthier and more densely populated than Mendocino and Lake Counties to the north. 

In reviewing the table, the income disparity is shocking - the median household income for Sonoma County is about 64% higher than neighboring Mendocino County.  Marin County, with its proximity to San Francisco, is the wealthiest county in terms of median household income and per capita income.  The 5-county average ratio of 1.8 between median household income and per capita income suggests that most households have two sources of income. Napa’s land area is slightly bigger than Marin County’s, but its economy is roughly half the size. Mendocino and Lake Counties are clearly rural with a population density that is a fraction of the other counties.  Mendocino County has twice as much land as the next biggest county, but the population is a fraction of the densities for Marin and Sonoma.

Moreover, Mendocino and Lake County are poor by California North Bay standards with a respective 16.1% and 16.5% of the two counties living under the poverty level compared to Napa (9%), Marin (7.8%), and Sonoma (9.1%).   The United States Department of Agriculture classifies any county with a poverty rate of more than 20% as “high poverty.”   The Dissimilarity Index and the GINI Index paint a similar picture. This dissimilarity index measures segregation in the counties; the GINI index measures the income disparity.

Dissimilarity Index for White vs Hispanic and GINI Index for Income

The initial reaction of a marketing professional might be to classify Marin, Sonoma, and Napa Counties as rich and Mendocino and Lake Counties as poor.   However, we will see that each of the North Bay counties has pockets of wealth and poverty.

This example focuses on the median home value as a measure of opportunity for measuring the wealth of the region.  Home value is often the single largest family investment and a measure of wealth. However, we learn about more than wealth; this data set is rich with other variables for study and provides detailed data to understand the inferences from Machine Learning Tools.

Consumer Market Analysis Using Unsupervised Learning

Unsupervised Learning, which includes Principal Components Analysis, provides statistics for reducing the “dimensions” of the data. This tool is especially adept at identifying common factors in datasets with thousands of variables without using labels — and therefore is considered unsupervised.

County-level data does not really give us a refined picture of the population and smaller regional economies.  Are there common factors for each county or are the counties completely different?  The American Community Survey provides a detailed breakdown of the social and economic microclimates in the counties. We can see from the two maps of Aggregate Income that the picture is complex.  Small businesses can take advantage of this knowledge by marketing to specific microclimates through targeted social media and other marketing channels.

The two maps below provide some insight into data available in Mendocino and Sonoma Counties.  The county subdivisions are the U.S. Census Bureau’s Census Tracts for organizing the Decennial Census.  These tracts can be further divided into Census Blocks for additional micro-region detail.

Note that the legend for Sonoma tops out at $400 million whereas the maximum for Mendocino County’s legend is $200 million.  One of the poorest regions in Sonoma County borders Mendocino County but is 2 times the income of Mendocino’s neighboring tract.

When we combine the two counties, we see there is an abrupt change in income levels on the borders of the counties, but there are also many areas of the counties that are similar. This section solves many of these puzzles using unsupervised machine learning to provide detailed insights about microeconomic climates that astute marketing managers use to tailor specific messages that connect with local populations.

Principal Component Analysis (PCA)

Benefit: PCA provides a tool to reduce the number of features (variables) that we need to consider while maintaining most of the information from those features.  As will be discussed in the next few pages, the component information provides deep insights into the key items that unite or separate populations.

This Principal Components Analysis identifies the factors that drive the demographics of the area. Principal Components are vectors of numbers used to reduce the number of features (variables) in analysis but still describe a census tract with great mathematical detail. Each component has factor loadings that further break down the variables associated with each factor. This can be useful for micro-economic research, as it can help to identify the key factors that drive economic activity in different tracts.

Eighty-six percent of the variance in median home values can be explained by the first 8 principal components of this dataset (PC1 to PC8). As explained above, each principal component provides the factor loadings for the variables.

Ranking of the most important Principal Components by contribution (the first 10 provide 92.42% of the information):

             Contribution                       Cumulative Contribution

PC1:       34.82%                                34.82%

               PC2:       23.41%                                58.23%

               PC3:       10.23%                                68.46%

               PC4:         5.58%                                74.03%

               PC5:        4.06%                                78.09%

               PC6:        3.42%                                81.51%

               PC7:        3.06%                                84.57%

               PC8:       2.96%                                 87.52%

               PC9:       2.64%                                 90.17%

               PC10:     2.25%                                 92.42%

When we look at the map of the factor loadings for each principal component, we begin to understand how they reduce the dimensions without losing the ability to model volatility.

 Principal Components for the North Bay Region

Each principal component has several factor loadings.  The factor loading is positive if the green bar juts to the right and negative if it juts to the left.  Each component is composed of various combinations of factor loadings or exposure to the variables – examples are:

               College Education

               Foreign Borne

               Renter Occupied Housing

               Population Density

               Median Age of the Structures

               Median Age of the Population

               Hispanic

               Asian

Principal Component 1, which explained 34.82% of the volatility in the data, is heavily positively loaded for the following key factors:

               White

               Total Population

               Living in the same house last year

               Owner occupied

               Median Income

               Higher Aggregate Income for the tract and by household

Principal Component 2 explains 23.41% of the volatility and contrasts strongly with #1:

               Populated areas (same as #1)

               Renting housing

               Low percent white

               High foreign-born

               Low owner-occupied housing

               Low income

               Highest Hispanic loading

Principal Component 3 explains 10.4% of the volatility:

               Negative exposure to White

              Most Positive Wages to Social Security

               Foreign Born

               High Percent College

               Negative weighting on Owning House

               Negative on Living in Same House Last Year

               Positive loading for Aggregate Income per Person

The Principal Components can then be used to construct a mathematical model of the census tract.

Waits for respective principal components

With Principal Components, marketing can develop very precise mathematical descriptions for target neighborhoods. The weights for each Principal Component are assigned to each tract and serve to mathematically characterize the location in detail.  For instance, the weight for PC1 for Covelo is -4.381 because it is the site of the Roundtree Indian Reservation, and the white population is a relatively small percentage of the population.  However, Covelo’s weighting for PC7, with the strong factor loading for Native Americans, is 6.850.  West Novato in Marin County, on the other hand, is a neighborhood with many whites, its weight for PC1 is 7.105, while the weight for PC7 is 0.295.  These two tracts contrast with East San Rafael with one of the highest exposures to PC2 that is heavily loaded for Renters and Hispanics and very few college graduates. This is only the beginning of insight into these tracts and the possible combinations provide deep insight into the demographics of the tract.

We can now map the importance of the component to each census tract. NOTE: This paper will only look at the top 3 principal components.

PC1 Loads are heavily influenced by factors associated with the white population (see the yellow and light green tracts):

PC2 Loads Hispanic and associated variables (Note that once again, tracts that have a high Hispanic contribution and are yellow-green):

               Positive:

Percent Foreign Born

                              Renter Occupied

                              Population Density

                              Hispanic

               Negative:

                              White

                              Percent College Graduate

PC3 Loads Wealth, foreign-born, college education, and negative for receipts of social security: PC3 is heavily influenced by income from wages, is the only factor where race is not a major loading, and tends to be more important in the South.

Principal Components Regression: Supervised Learning Applied to Unsupervised Learning Results

Benefit: Principal Components Regression provides another view of the data, like looking at a house from the front and then walking to view from the side.

The previous PCA focused on component-by-component analysis.  The PCA regression gives a tool to incorporate all the components in one equation to evaluate a tract.  Note: The PCs can be used as indices and equations can be used to develop a score for each tract.

There is more to the Principal Components story. Principal Components can be used for principal components regression, in which the derived components themselves are used as model predictors. Generally, components should be chosen that account for at least 90 percent of the original variance in the predictors, though this will often be up to the discretion of the analyst. In the example below, we will fit a model using the first six principal components which represent 80% of the model variance and the outcome variable is once again the log of the median home value.

Principal Components Regression

Principal Components Regression Analysis

With an R-squared value of 71.46%, the model fit is close to the first regression model fit of 75.05% earlier in this paper. The PCA model is also statistically significant. We can think of principal components as indices that measure the economic activity in the region.  The advantage of this analysis is that we can also examine the contributions of the factors to each census tract based on the factor loadings. 

Table of Selected Observations from Map

This regression provides an economic index of the well-being of a census tract.  The average is the intercept, 13.42.  One of the higher scores (the score is calculated by multiplying the Estimates by the factor loading for each Principal Component) is 14.81 for Tiburon in Marin County with a Poverty Rate of 5%.  One of the lowest scores is 12.35 for Kelseyville in Lake County with a Poverty Rate of 21%. 

The regression scores are based on the following combination of variables.

PC1:  A strong positive contribution to the median value of housing for the entire region.

PC2: Negative factor in housing valuation and the second most significant variable.

PC3: As noted above, this component measures the factors associated with high income in a region. It makes sense that this component would have the highest estimate (0.1615) and would be the most statistically significant (t value of 15.864).

PC4: This component loads heavily for a high percentage of owner-occupied houses, low number of renters, low percentage White, high percentage Asian, large number of rooms in the house, and occupied by the same person last year.

PC 5: Not significant (small estimate and low t value).

PC 6: This factor loads heavily for residents of Pacific Island Descent and most of the locations are in Napa Valley.  It also has a heavy factor loading for the age of the structure and a negative loading for Other Races and Hispanics.

PC 7: There are two positive main loadings for this component: percentage Native American and percentage Black.

PC 8: There are two positive main loadings for this component:  median structure age and other race..

PC 9: There are three positive main loadings for this component: negative exposure to Pacific islanders, other races, and Native Americans.

Supervised Learning: Geographically Weighted Regression

Benefit:  The linear regressions estimate global relationships between the dependent variable (variable being predicted) and the independent variables (used to predict).  Per Walker, “This lends itself to conclusions like ‘In the Dallas-Fort Worth metropolitan area, higher levels of educational attainment are associated with higher median home values.’ However, metropolitan regions like Dallas-Fort Worth are diverse and multifaceted. It is possible that a relationship between a predictor and the outcome variable that is observed for the entire region on average may vary significantly from neighborhood to neighborhood. This type of phenomenon is called spatial non-stationarity, and can be explored with geographically weighted regression, or GWR (Brunsdon, Fotheringham, and Charlton 1996).”

In the following analysis, we map the Median Home Value and then compare that to the local R-squared to find the local variations from the global conclusions that we reached using PCA and PC Regression.

Geographically Weighted Regression Map

The below shows the predicted values for the log of the Median Home Value and the results are very similar to the Principal Component Analysis.

With the base R-squared of 75% on the legend, this map shows how the R2 deviates by census tract. The model performs well across the region but is better in some of the more rural areas of the region, especially in the very south, the eastern, and northern census tracts where the R-squared ranges from 80% to above 90%. Note: the model deviates most in the rural regions of Mendocino and Lake Counties.

The map below shows the relationships of Percent of Owner-Occupied Housing in local tracts to the overall model. Recall that the relationship in the percentage of Owner-Occupied Housing (OOH) to home value is negative for the region. The dark purple areas on the map are those areas where the global relationship in the model reflects the local relationship, as local parameter estimates are negative. The areas that stand out include the high-density area of lower Marin County, where median home values are very high. However, in the mostly northern rural tracts of the region, the estimate is zero indicating that the local percentage of owner-occupied housing does not affect the value.

The population density parameter estimate was positive for the entire equation. The tracts in Marin County in the south, in southern Mendocino County in the center, and Lake County in the central east have no local beta. Once again, the key wine-growing regions in Napa and Sonoma go against the overall trend — the property values increase with lower population density.

Cluster Analysis: Unsupervised Learnin

Benefit: Cluster Analysis identifies economic characteristics that explain the spatial distribution of economic activity and groups the tracts into data sets or clusters with similar characteristics versus clusters that differ significantly. Cluster analysis on PCs provides insights into economic opportunities in micro-regions. Does the tract present qualities for economic growth or is it characterized by a low-income or impoverished economy?

While Cluster Analysis can be run on raw data, many data scientists apply PCA to the data before analyzing the data with clustering algorithms.  This two-step procedure reduces the noise in the clustering results. While PCA and Cluster Analysis are similar, the two techniques have different goals.  If a study has 100 features (variables), PCA tries to condense that information into a smaller number of features that really matter.  Cluster Analysis, on the other hand, seeks to represent the 100 features into a set of clusters that are internally the same but significantly different from the other clusters.

After several iterations, I found that 6 clusters provided the best fit and separated the PCs into distinct groups.

In review of the factor loadings in the dot plots from previous sections, PC 1 is the component that represents the White population with higher income, some college graduates, and owner-occupied housing.  PC2 represented a mostly Hispanic population that was dominated by renters.  The dots represent the tracts, and the color identifies the cluster assigned.  Tracts that are to the right of zero on the horizontal axis are weighted positively toward PC1 as with Clusters 2, 3 and 5.   Tracts that are above zero on the vertical axis are weighted to PC2; Clusters 1 and 4.

Plotting PC1 against PC3 (income - no race component) shows that there are several distinct income groups. Cluster 1 (Hispanic, renter in the denser south region) ranks positively in PC3 as does Cluster 3 (Rural, White, Small Towns) and Cluster 5 (Wealthier denser populations in the south). Cluster 2 (Rural population centers with mixed PC1 and PC2), Cluster 4, and Cluster 6 do not rank as high in wealth.

Cluster #1(Red): Rural areas, firmly Hispanic (PC2) with few Whites (PC1). In general, these regions are in pockets that lie between bigger regions, like the 101 corridor from Cloverdale to Healdsburg, a small tract on the southeast side of Santa Rosa, and in Napa Valley.

Cluster #2(Blue): Cluster 2 includes heavily commercial areas in the south that have positive exposure to both PC1 and PC2, where both White and Hispanic residents are strong.  In the south, this represents the Highway 101 corridor.  Most of these areas have the densest populations in the region and are significantly more populated than the surrounding census tracts in the region.  The big blue tracts in the north are Willits, Ukiah, Kelseyville, and Cloverdale.

Cluster #3(Green): Represents agricultural and wine-growing regions with positive exposure to PC1 (White, owner-occupied).  The green areas are northern Santa Rosa to Healdsburg regions and include some of the Russian River wine region and key wine regions of Napa.  Like Cluster 5, this cluster has no exposure to PC2 – the Hispanic-dominated Principal Component.

Cluster #4(Purple): Represents the poorer mixed races in rural, and agricultural regions.

Cluster #5(Orange): This cluster has a negative weighting for PC1 and for PC2, meaning the residents are predominantly white and in the higher-income areas of Marin and Sonoma County.

Cluster #6(Yellow): Equally PC1 and PC2, but low exposure to PC3 (wealth component). This factor covers the Covelo tract in the northeastern corner, some of the poorer neighborhoods around Clear Lake in Lake County, and the Point Arena and Navarro/Boonville regions of western Mendocino County. These areas are sparsely populated and either agricultural or heavily forested.

Summary

While this study delivered some deep insights into the demographic breakdown of the North Bay Region, it is a preliminary case study or a first step that small and medium-sized companies can take to understand customers and improve customer experience.

AI and Machine Learning Tools can spearhead an effective defense against bigger companies and competition from new, disruptive technology.  Despite the length of this paper, it only addressed a small group of customer preferences, and, in many ways, it raises as many questions as it answers.

Some of the solutions that the analysis highlighted:

1.      A financial company might want to advertise the highest interest rates on deposits to the wealthy, older communities in Marin.

2.      Send out advertisements in Spanish to the heavily Hispanic Communities.

3.      Promote community programs in the poorer tracts of Lake County and Mendocino

4.      A finance company might want to promote home equity loans in regions with high home ownership

5.      On the other hand, the same finance company would promote affordable home loan programs in Spanish to tracts with a large percentage of the Hispanic population who rent.

As the paper shows, AI and Machine Learning give Small and Medium-Sized businesses the tools to counter disruptive market developments with a deep understanding of their market and an intense commitment to improving the customer experience, from a seamless delivery of products to attention to customers’ specific needs.

In the bigger picture, the ability to use AI also depends on a company’s corporate structure. The modern firm needs to transform itself into a digital, agile organization that can share AI throughout the firm to survive the coming market disruptions.

Neural Profit Engines provides a suite of chief financial officer services under the brand name Neural Financial Officer. 

·       Big Data studies to aid planning and financial analysis

·       Business strategy based on AI and Machine Learning

·       Data cleaning - labeling and preparation

·       AI and Machine Learning analysis of big data and trends

·       Planning and financial analysis

·       Bookkeeping and accounting services

·       Company training for ChatGPT and Bard

Vernon H. Budinger, CFA

Chief Executive Officer

vernon@neuralprofitengines.com

www.neuralprofitengines.com

+1(707) 513-0880

The Facts about Raging Inflation and a Possible Recession: Usable Information for Small Businesses

The July 13th announcement that inflation climbed to 9.1% seems to have stunned many in the finance industry. As a Wall Street veteran and a Finance Professional for over 40 years, I am surprised at the level of confusion regarding the state of the economy, the dangers of inflation and the probability of recession.   

The major driver of the confusion has been Mainstream Media’s politicization of the economic story; MSM is selling the Biden Administration’s story instead of honestly reporting the factors driving the U.S. economy. Then there are the Nobel Prize-winning economists (Paul Krugman for one) who mislead the public and journalists to promote their politics.  This blog presents the facts and logic so that small business owners, who are struggling with inflation and horrified at the chances of a probable recession, can understand the U.S. economy and evaluate management options for their small businesses. 

Note: to keep this blog at a reasonable length, I have left out the debate on the Biden Administration’s oil policy because it is not necessary to demonstrate that government policies are responsible for the dire economic scenario.

Three recent economic studies in the past 2 months directly address the key issues driving inflation and a probable recession.  The first, from the Federal Reserve of San Francisco, directly links the Biden Administration’s aggressive fiscal stimulus to the current persistent inflationary trends.  This ends the greedy corporation and evil Putin gaslighting.

The second study, a working paper for the National Bureau of Economic Research coauthored by Harvard University’s Lawrence Summers, concludes that the inflationary pain felt by today’s consumers is equivalent to the damage done in the late 1970s and early 1980s. This study clearly demonstrates that Federal Reserve Chair Jerome Powell and the current Fed governors will need to administer some strong economic medicine given that Fed Chair Paul Volcker used extreme interest rate hikes in 1980 and 1981 to exorcise inflation and send the economy into a deep recession.

Third, the University of Michigan released a gloomy Consumer Sentiment Index for June 2022 that shows that U.S. consumers are becoming very wary and drastically curtailing spending plans.  This is bad news for the GDP equation.

Paper #1: Federal Reserve of San Francisco - “Why Is U.S. Inflation Higher than in Other Countries?”

The S.F. Fed paper, authored by Oscar Jorda, Celeste Liu, Fernanda Nechio, and Fabian Rivera-Reyes, unites basic economic theory and sound statistical techniques (from previous peer-reviewed papers) to prove that the aggressively stimulative U.S. Government fiscal policy financed by almost uncontrolled printing of money is driving today’s inflation spiral.  The paper leaves no doubt that inflation is going to be persistent and difficult to correct with only monetary policy.

https://www.frbsf.org/wp-content/uploads/sites/4/el2022-07.pdf

To set the stage for the SF Fed’s paper, the graph on the next page illustrates the inflation talking points that most economists and journalists use for their analysis.   The key is the grey line showing U.S. Gross Domestic Product (GDP).  This graph clearly shows the correlation between GDP growth, the growth in U.S. Federal Debt (top blue line), growth in the U.S money supply (orange), and the U.S. Federal Government Spending (bottom yellow line). While this graph seems to clearly illustrate the forces driving the economy, and therefore inflation, the relationships are correlation not causation. However, it does show that the dramatic increase in government spending and federal debt were financed by printing money.

Source: St. Louis Federal Reserve Fred System

Note: Money = M2 = currency, checkable deposits, traveler’s checks and savings deposits

A link to inflation needs to tie GDP growth from government stimulus to inflation. The SF Fed paper uses an innovative study to determine causality as explained in the following quote:

Though many of the pandemic distortions are common to other countries, we show that U.S. inflation has risen more quickly and increasingly diverged from inflation in other OECD (Organization for Economic Cooperation and Development) countries.

First, the authors show that inflation is clearly much higher in the U.S. than in other OECD countries. True, some OECD countries are experiencing U.S.-like inflation, however, the authors address this in the second section of the paper.

The authors then show that U.S. direct fiscal transfers were also higher than the average OECD country. (Note: I use “fiscal transfers” or “fiscal stimulus” to refer to U.S. Government spending and “monetary stimulus” to refer to Federal Reserve operations.) 

Rather than trying to track and sum all the various fiscal stimulus programs adopted round the world, the authors directly measured disposable income in each country. Many journalists and politicians have made the argument that the U.S. fiscal stimulus - $1,200 per person and $500 per child plus $600 a week for unemployment – was not enough to substantially change the finances of U.S. consumers.  However, the SF Fed data shows that this was not true.  This was substantial fiscal stimulus not seen on average in other OECD countries.  The two peaks in the U.S. data “reflect the CARES Act, signed into law on March 27, 2020, and the American Rescue Plan (ARP) Act of 2021.”

The key connection: “Did excess disposable income drive inflation?”

The authors then applied the Phillips Curve analysis to the data. Phillips Curves express inflation as a function of the unemployment rate. As the authors explain: “inflation reflects a combination of the public’s views on future inflation, inflation inertia, and how hot the economy is running.  Because the array of policy measures introduced during the pandemic to counterbalance the economic effects of lockdowns, common labor market statistics, such as the unemployment gap, are not reliable.”

The authors designed an ingenious method to measure the Phillips Curve effect: they compared inflation in countries that used aggressive stimulus measures to inflation in countries that were more passive.  “Using the Phillips Curve logic, we can reasonably compute the effect of pandemic support of measures on the inflation forecast. The idea is to compare the countries that, like the United States, introduced aggressive support measures, which we call the policy ‘active’ group, versus the less aggressive, or policy passive group before and after the pandemic.”   

Figure 3 clearly shows that inflation would have been much lower without the aggressive stimulus. The authors use Core CPI (without food and energy), which clocked in at roughly 5% at the time of their study. They predicted that the Core Inflation rate would have been roughly 2% without the stimulus.

The green shaded area displays the uncertainty around their statistical forecast.  Think of the green area as a series of bell curves with the maximum for each curve located over the green line.  The shaded area grows because the forecast becomes more uncertain as it is used to predict farther ahead in time.  Bottom line: the study shows that there is strong evidence that excessive fiscal stimulus drove the elevated inflation experienced in “aggressive” countries.

To address the persistence of today’s inflation, allow me to sneak one more SF Fed paper into the mix.  This paper, titled “Untangling Persistent versus Transitory Shocks to Inflation,” measured the persistence of inflation relative to the transitory factors.

https://www.frbsf.org/economic-research/publications/economic-letter/2022/may/untangling-persistent-versus-transitory-shocks-to-inflation/

Quote:

At the end of the data sample in March 2022, the shock volatility ratio is around 2, which means that persistent shocks to inflation are about twice as volatile as transitory shocks. This result implies that persistent shocks are the more important driver of recent inflation movements. The higher shock volatility ratio in recent data also implies that the optimal forecast should place more weight on recent elevated inflation readings, reflecting an increased likelihood that longer-run inflation has drifted up. But it is important to note that the FOMC’s decision in March 2022 to begin an “appropriate firming in the stance of monetary policy” would be expected to influence the future behavior of inflation.

 In other words, inflation is driven mainly by government spending, and it is persistent – not transitory.

Paper #2: “Comparing Past and Present Inflation,” Marijin A. Bolhuis, Judd N. L. Cramer, and Lawrence Summers.

https://www.nber.org/system/files/working_papers/w30116/w30116.pdf

The significance of this paper cannot be overstated.  First, Lawrence Summers is the Charles Elliot University Professor at Harvard.  Second, he is not a right-wing Fox News or Breitbart shill.  Summers is a true-blue Democrat. He was the Secretary of Treasury under Bill Clinton and the Director of the National Economic Council for Barak Obama. If Summers is warning about inflation, we should listen.

The paper’s introduction states:

As concerns about US inflation have grown, the Consumer Price Index (CPI) has come under closer scrutiny. The CPI grew 8.3 percent in the twelve months ending in April, down slightly from the previous month but still well above any other period since 1981. While a worrying figure, this remains far below the official March 1980 peak of 14.8 percent. That the headline number had already fallen to 2.5 percent by July 1983, following the policy decisions of Federal Reserve Board Chairman Paul Volcker, has served as the exemplum of the power of hawkish monetary policy (Goodfriend and King 2005). Since much less of a decline is needed to return to trend today, some commenters have suggested that policymakers might be able to decrease inflation towards desired levels without large macroeconomic consequences (DeLong 2022; Krugman 2022). Yet, methodological changes in the CPI over time make drawing conclusions from these types of intertemporal comparisons fraught.

The author’s findings state:

Our estimates suggest that the current inflation rate is closer to the peak of other cycles than the official CPI data suggest. …  We draw two sets of conclusions. First, our observations imply that the current inflation regime is closer to that of the late 1970s than it may at first appear. In particular, the rate of CPI disinflation engineered in the Volcker-era is significantly less when measured using today’s treatment of housing. In order to return to 2 percent core CPI today, we need nearly the same 5 percentage points of disinflation that Volcker achieved.

Note that even Larry Summers is calling out Krugman for gaslighting the American public in an NBER paper!  However, let’s focus on these frightening findings and put this work in perspective with a historic graph of interest rates and inflation.

To make sure the reader understands the importance of Summers, et. al., they are reporting that the consumer pain felt in 1980 and 1981 (when inflation was at 14%) is EQUAL to the consumer pain felt now because the 1981 inflation data needs to be adjusted for the basket of goods and services in 2022 and computational changes in measuring housing.  Once this is done, the authors conclude that the current inflation is about the same as in 1980.

As we can see from the graph above, the Fed Funds Rate, the main Federal Reserve tool for controlling economic activity, peaked at 19.1% in July 1981; inflation had peaked over a year earlier in March 1980 at 14.8%.  Basically, Volcker and the Fed needed to increase interest rates ABOVE the inflation rate for an extended period to control inflation. 

The reason? Before the Fed took control, investors could borrow at interest rates below inflation – that means that they were being paid to borrow since they were paying back with dollars that had devalued more than the interest rate they paid – in other words, they were paid to borrow.  Then investors used funds from the loans to invest in fixed assets that appreciated at the inflation rate.  This is the arbitrage that wealthy families used in Germany in the 1920s hyperinflation to become incredibly wealthy.  Savvy investors are doing the same today, that is one reason that the real estate market continues with strong positive performance.

Look at the current relationship between inflation and the Fed Funds Rate – the Consumer Price Index clocked in at 8.3% in May 2022 and updated Fed Funds are sitting at 1.5% to 1.75% (this paper was being released just as the Department of Labor released the 9.1% number for June).  As Summers and crew explain, the Fed will need to inflict some strict monetary tightening and nasty economic medicine to control the current economy.  However, the economy is not expecting the Fed to increase interest rates above 4.0%.  If 1981 serves as a guide, the Fed will need to push rates to 8% or 9%.

The risk to my 9% Fed Funds forecast is the possibility that the Fed does not need to raise interest rates above inflation because of “quantitative easing” tools.  In previous economic downturns, the Fed relied on the overnight rate to steer the economy.  However, during the Great Recession of 2008, the Fed started practicing “quantitative easing”; they bought market securities with longer maturities to inject liquidity into the economy.  This tool was especially effective in 2008 and 2020 when short-term interest rates dropped to almost zero.  Similarly, the Fed could tighten liquidity by selling these longer maturity securities.  However, that would still remove liquidity from the economy, slow economic growth, and risk a recession.

Bottom line, there is no doubt that the U.S. Federal Reserve will need to administer more monetary tightening than the market is expecting.

Paper #3: Michigan Index of Consumer Sentiment Drop Ties the Lowest Levels Ever Measured

To understand the importance of the Michigan Index of Consumer Sentiment – a graph that you are about to see - you need to understand this Gross Domestic Product (GDP) equation.

GDP = Consumption + Business Investment + Government Spending + Net Exports

Terms:                                               Contribution to GDP

C = Consumer Spending                           70%   

 I = Business Investment                         18%

G = Government Spending                      17%     

E = Export minus M = Import                  -5%

Note that Consumer Spending contributes a whopping 70% to the GDP.  When you see the Consumer Sentiment Index drop dramatically as seen in the chart below, then you know that that our economy faces some difficult times since it has nearly 4 times the weight of the next most important component of GDP. 

Source: University of Michigan Survey of Consumer Confidence

This is the statement from the University of Michigan:

The final June reading confirmed the early-June decline in consumer sentiment, settling 0.2 Index points below the preliminary reading and 14.4% below May for the lowest reading on record. Consumers across income, age, education, geographic region, political affiliation, stockholding and homeownership status all posted large declines.”

Inflation continued to be of paramount concern to consumers: 47% of consumers blamed inflation for eroding their living standards, just one point shy of the all-time high reached during the Great Recession.

Many economists and journalists react to these numbers with the hope that consumers will reach into their savings accounts and spend down savings.  Oh yeah, I haven’t posted the trend for the Personal Savings Rate, have I?

This is probably why consumers are so scared; they have already drawn down their savings.  This Personal Savings Rate graph is what led Jamie Dimon, CEO of J.P. Morgan, to predict a severe recession (an economic hurricane). We can see the two income spikes that were in the SF Fed study; the Personal Savings rate reached an unheard-of level of 33.8% in April 2020 and spiked again to roughly 26.6% in March 2021.  J.P. Morgan reported that the average personal checking account held $1,500 at the end of 2019 and that had risen to $7,500 in 2021. Now personal savings have plunged to 5.4% in May 2022.  The low was 2.1% in July 2005, so consumers have some room to go, but not much.

Now, when we consider the rest of the GDP equation, Business Investment is the only variable that has been contributing to positive economic growth.  This could change with the increases in interest rates – we have seen the fallout in the Cryptocurrency market from Crypto firms that were dependent on zero interest rate policies.  Hopefully, Crypto problems will not spread to other business sectors.

Government spending is also falling, as can be seen in the first chart.  The government is spending less as the spending for employment supplements and other COVID stimulus packages expires.

Additionally, Net Exports are also likely to be a drag on GDP growth as the Fed pushes up interest rates, the dollar has strengthened and will continue to strengthen (it just traded at par with the Euro).  This makes exports more expensive and imports cheaper, which leads to a growing trade gap and reduces GDP.    All in all, it looks like a perfect storm for the economy.  Did we mention the war in Ukraine?

Inverting Yield Curve

This is a bonus section (since it has been in the headlines recently) because an inverted yield curve is a very reliable predictor of inflation.  Investors subtract the 2 Year Treasury Yield from the 10 Year Treasury Yield to calculate the slope of the yield curve.  As of Thursday, July 14, 2022, the 2 Year was yielding 3.12% while the 10 Year yielded 2.94% for a negative 18 basis point curve inversion.  While this is not a good sign, an inversion of more than 15 basis points has a 100% accuracy record in prediction recessions.  (Recessions are the gray bars and the curve is inverted when it dips below the black horizontal line.)

Conclusions and Implications for Small Businesses

Inflation will continue at elevated levels into the foreseeable future and imperil U.S. economic growth.  This is not transitory; inflation will persist even if the inflation-adjusted economy stops growing - stagflation.  Inflation alone is enough to threaten economic growth as consumers curb spending because they need to spend more of their fixed budget on necessities.  However, the Fed’s economic medicine will most likely be more damaging since the Fed has already lost control of the economy.  The Fed will need to become much more aggressive and we are not talking about exact science with precise doses of economic medicine that will allow the Fed to safely land this economy at 2% inflation.

Conclusion 1:

Inflation is the result of financing aggressive government stimulus with growth in the money supply – printing money.  This inflation was expected, is not transitory, is not the result of the war in Ukraine, and will persist for the foreseeable future. The US printed money to finance this spending and the easy monetary policy financed the nearly uncontrolled spending – now we are paying the piper.  While monetary growth has fallen from the 12% year-over-year gains seen in 2021, it is still above average for the U.S. economy.

Implication 1 for Small Businesses:

1.      Accept inflation as the new normal for now.  Become aggressive in learning to deal with inflation.

2.      Understand the importance of pricing and expense control. 

3.      With respect to pricing, ignore the profits reported in GAAP accounting.  Prices need to be adjusted now to reflect future price increases for inventory, not when you buy inventory or equipment in the future.

Conclusion 2:

The level of inflation is as economically painful as it has ever been in modern financial history.  Those who lived through the late 1970 and early 1980s can tell you that it was a very difficult period. We are currently going through the same pain and the economic medicine needed to bring inflation down to the 2% long-term target will taste nasty.

Implication 2 for Small Business:

For the second time - get moving if you are postponing action to address inflation! Most of the probability distribution around future inflation is on the upside, as can be seen in Figure 3 from the Federal Reserve of San Francisco paper.  This means that the inflationary trend will most likely continue and there is little chance that inflation will drop dramatically in the short term.

1.      Look at your pricing options; where can you safely increase prices?

2.      Maintain good relationships with your clients, especially your biggest clients. 

3.      Use social media to communicate with your clients and potential clients.

Conclusion 3:

There will most likely be a recession and it could be one of the deeper recessions.  While the Fed correctly points out that jobs are still strong and the economy is still pressing ahead, administering the monetary medicine needed to bring inflation to 2% will be difficult and the financial markets are underestimating the doses needed to slow economic growth.

Implications for Small Business:

1.      Go for price increases now where possible.  Don’t wait for the economy to collapse.

2.      Understand your cash flow – profits don’t pay bills.  Cash pays bills. The first thing that finance and investment professionals do is tear apart the profit and loss statement to understand the future cash flow picture.

3.      Talk to your bank about a line of credit or other financings now when you don’t need it. If you can SAFELY (meaning that the amount represents a risk-adjusted small portion of your business) borrow money at a 1- or 2-year rate now to purchase inventory or other assets - buy now before prices increase further and pay the money back with dollars that are worth 5% to 10% less than today.

4.      Look at expenses to bring them in line with revenues.

5.      Eliminate unprofitable or low-margin product lines and introduce new products that diversify your income stream.

Check out the Neural Profit Engines website for more papers regarding the dangers of combining GAAP accounting and inflation.  Contact us for more insights into positioning your company to thrive in today’s chaotic business environment.

Vernon Hamilton Budinger, CFA and CAIA

Neural Profit Engines

Position Your Company to Survive through Grow

www.neuralprofitengines.com

vernon@neuralprofitengines.com