I break down the key components of Powell’s observations (using the transcript from the CNBC TV recoding of the live event) and then juxtapose them with the prospects for inflation going forward.

The S&P 500 (SPY) since the October lows overlayed with my past short-term trading calls. (Source: Tradingview.com)

Is Policy Really Restrictive?

Powell observed that “the labor market is about as tight as it was before the pandemic in 2019, and that’s good. Real wages are now positive, and unemployment has been below 4% for 27 consecutive months, something that hasn’t happened in half a century. So, the labor market is very strong, with signs of gradual cooling and rebalancing, so that the supply and demand for workers are coming back into balance, just as we would want.” The logic did not make sense at first. Powell provided the evidence of balance in the labor market later on in the discussion. He noted that “the ratio of vacancies to unemployed people, for example, has gone from more than 2 to 1 to now 1.3 to 1, and it was 1.2 before…You also see wages coming down, and you see quits coming down.” Most importantly, Powell used these observations as evidence that “policy is probably restrictive.”

Powell went on to add that restrictive policy is weighing on rate-sensitive spending. However, the conventional indicators of financial conditions still show they are just as loose as they were before the Fed started hiking, January 28, 2022 to be exact. In other words, either the market has done some serious advanced front-running of rate cuts or current policy is not nearly as restrictive as the Fed thinks. Even the stock market’s all-time highs suggest monetary conditions are not tight. From the Federal Reserve Board of Chicago’s National Financial Conditions Index (NFCI) (follow the black line for ANFCI):

Even the hand-wringing this year from the Federal Reserve about resilient inflation implies that policy may not be restrictive enough (whether that means rate hikes, or, more likely, extending the time horizon for current rates).

Resilient Inflation

Powell repeated a familiar refrain about inflation not falling as fast as expected given presumed restrictive policy. Thus, Powell had to show deference to the possibility that today’s high rates will need to stay in place longer than expected:

“The first quarter in the United States was notable for its lack of further progress on inflation. We had higher readings in the first quarter, higher than we expected. We did not expect this to be a smooth road, but these were higher than I think anybody expected. So, what that has told us is that we’ll need to be patient and let restrictive policy do its work.”

This comment tells me rate cuts are nowhere on the horizon. Moreover, Powell noted “we have the highest interest rates in some time. It may be that it takes longer than expected to do its work and bring inflation down.”

Most telling were Powell’s words when a question offered the opportunity to reassure the market that rate cuts were still on the way. Emphasis mine:

“So, as I just mentioned, I do think it’s really a question of keeping policy at the current rate for a longer time than had been thought. By many measures, the policy rate is restrictive. The question is, is it sufficiently restrictive? And I think that’s going to be a question that time will have to tell. Entertain the possibility? That could be a very small probability, but I have said that I don’t think it’s likely based on the data that we have that the next move we make would be a rate hike. I think it’s more likely that we’ll be at a place where we hold the policy rate where it is.”

Note what Powell did NOT say: it’s more likely that rates will be lower by the end of the year (or even next year).

So, nowhere in the discussion did Powell provide a line for the market expect rate cuts anytime soon. Thus, he has primed the market to get excited about any soft data that could provide the seeds of an excuse for the Fed to cut rates. Powell’s positioning is made even more potent by the stock market’s ability to trade at all-time highs despite all the hand-wringing about inflation this year.

Inflation Forecast

Powell expressed a lot of uncertainty about the path for inflation despite his confidence that “we will do that, that we will get inflation down to 2%.” I am not sure how rate cuts help bring down inflation (Turkey sure learned a central bank cannot cut rates to lower inflation), so I have to assume Powell has an image of “higher for longer” monetary policy rates.

Responding to a question on whether inflation could end up being more persistent than expected, Powell said: “I think we need more than a quarter’s worth of data to really make a judgment on that.” It’s like the perpetual data delay. It will be interesting to hear whether April’s inflation data finally gives the Fed more confidence that inflation is coming down sustainably. Time is running out for the Fed to start telegraphing the first rate cut, so the pressure is on the Fed to spin a story about April and May inflation that points to the 2% target.

When I look a the core CPI graph (year-over-year percentage change in the price index), it looks like inflation will soon resume a steep descent. To apply a modicum of extra rigor, I ran a simplistic forecast through ChatPGT’s Data Analyst. I let the model pick its forecasting method. This is a very simplistic forecast because it only uses the aggregate CPI time series. A more accurate forecast would consist of aggregating the forecasts of individual components of core CPI.

I provided the model core CPI (measured year-over-year) data back to 2016. I wanted the model to surmise that “equilibrium” consists of a tight range for inflation. I did not provide any more data because I figured the further away history gets from the pandemic era, the less likely the model has any value (another downside of using a single, aggregated time series for the forecast).

The chart below combines the actual core CPI data with the forecast. The “funnel” around the forecast shows the confidence interval within which inflation will fall in the future at 95% certainty. The wide swath of possibilities is consistent with Powell’s uncertainty about the path ahead. The flat average forecast is a typical behavior of time series forecasting where there are no definitive seasonal effects detected or modeled future episodic events. (For you ChatGPT and statistics wonks, I provide an Appendix below where ChatGPT explains the model and justifies its approach).

Source: U.S. Bureau of Labor Statistics, Consumer Price Index for All Urban Consumers: All Items Less Food and Energy in U.S. City Average [CPILFESL], retrieved from FRED, Federal Reserve Bank of St. Louis; May 15, 2024.

The Bond Market Rules

Putting aside vain attempts to forecast future inflation, the bond market is the ultimate arbiter of inflation expectations and/or how inflation may impact interest rates. The bond market in the form of the iShares 20+ Year Treasury Bond ETF is at a very critical (technical) juncture. The downtrend in TLT (uptrend in rates) was the ultimate guide in pushing the market into its April correction. The rebound in stocks for May perfectly correlates with the rebound in TLT (lower rates). The “soft” April CPI report sent TLT up 1.4% and right up against the top of the downtrend channel.

iShares 20+ Year Treasury Bond ETF (Source: TradingView.com)

A confirmed breakout with a higher close should open the floodgates of bullishness for the stock market. A failure at resistance from the upper part of the channel could trigger a false breakout for the indices. Thus, given this year’s correlation with stocks, I reloaded on TLT puts as a hedge on my long trading positions.

Overall, the lesson in all these dynamics is that inflation combined with the interest rate response should continue to dominate the market’s underlying narrative. With rates on hold, even the Fed is just along for the ride for now.

Be careful out there!

Appendix (ChatGPT explains and justifies the CPI forecast)

{the following is verbatim including ChatGPT’s musings on data types and fixing its own processing problems}

The data consists of monthly core CPI year-over-year percentage changes starting from January 2016. To project the CPI for the next 24 months, I will use the ARIMA (AutoRegressive Integrated Moving Average) model, which is well-suited for time series forecasting.

Justification for ARIMA Model:

1. Handling Trends and Seasonality: ARIMA can model both trend and seasonality in the data, which is crucial for accurate CPI forecasting.
2. Flexibility: The model can be adjusted to different time series characteristics through its parameters.
3. Proven Effectiveness: ARIMA has been widely used and tested in economic and financial forecasting, making it a reliable choice.

Steps for the Analysis:

1. Data Preparation: Convert the date column to a datetime format and set it as the index.
2. Stationarity Check: Perform tests to check if the time series is stationary. If not, apply differencing.
3. Model Identification: Use autocorrelation (ACF) and partial autocorrelation (PACF) plots to identify the order of the ARIMA model.
4. Model Estimation: Fit the ARIMA model to the data.
5. Forecasting: Generate forecasts for the next 24 months.
6. Confidence Intervals: Calculate confidence intervals for the forecasts.
7. Visualization: Plot the actual data, the forecasted values, and the confidence intervals.

Let’s start with the data preparation and initial analysis. ​​

The Augmented Dickey-Fuller (ADF) test results are as follows:

• ADF Statistic: -0.988
• p-value: 0.758
• Critical Values:
  • 1%: -3.508
  • 5%: -2.895
  • 10%: -2.585

Since the p-value is greater than 0.05, we fail to reject the null hypothesis that the time series is non-stationary. This indicates that the series needs to be differenced to make it stationary.

Let’s apply first-order differencing and recheck for stationarity. ​​

The ADF test results after first-order differencing are as follows:

• ADF Statistic: -2.447
• p-value: 0.129
• Critical Values:
  • 1%: -3.509
  • 5%: -2.896
  • 10%: -2.585

The p-value is still greater than 0.05, indicating the time series is not yet stationary. Let’s apply second-order differencing. ​​

The ADF test results after second-order differencing are as follows:

• ADF Statistic: -2.843
• p-value: 0.052
• Critical Values:
  • 1%: -3.509
  • 5%: -2.896
  • 10%: -2.585

The p-value is now close to 0.05, indicating the time series is approaching stationarity. Since it’s marginally above the threshold, we will proceed with ARIMA modeling with second-order differencing.

Model Identification

Next, we’ll use ACF and PACF plots to identify the appropriate orders (p, d, q) for the ARIMA model. ​​

The ACF and PACF plots provide insights into the appropriate orders for the ARIMA model:

• ACF (Autocorrelation Function): The ACF plot shows significant lags at various points, suggesting that an MA component (q) might be needed.
• PACF (Partial Autocorrelation Function): The PACF plot shows significant lags at the beginning, suggesting that an AR component (p) might be necessary.

Given the plots, an ARIMA model with orders (p, d, q) will be selected. Based on visual inspection, we can start with ARIMA(2, 2, 2) since both ACF and PACF show significance up to lag 2.

Model Estimation

Let’s fit the ARIMA(2, 2, 2) model to the data and generate forecasts for the next 24 months. ​​

It seems there’s an issue with the data types in the forecast confidence intervals, which might be causing the error. Let’s inspect the forecasted values and the confidence intervals to identify the problem. ​​

The forecasted values and confidence intervals appear to be valid. The issue might be due to incorrect data handling during the plotting process.

To resolve this, let’s explicitly convert the confidence intervals to a numeric type before plotting. ​​

It appears the error persists despite converting the confidence intervals to float. This might be due to a mismatch in the indices of the forecast intervals and the forecasted mean.

To resolve this, I’ll ensure the indices of the confidence intervals match those of the forecasted mean before plotting. ​​

The issue may be due to an incompatibility between the data types of the forecast index and the confidence intervals. To resolve this, I will ensure all data is in the correct numeric format and use the correct plotting function.

I’ll convert all relevant data to the same data type and then re-plot. ​​

The plot now correctly displays the actual data, the forecast, and the confidence intervals.

Summary of Analysis and Forecast:

• Data Preparation: The time series data was checked for stationarity and transformed using differencing to achieve stationarity.
• Model Selection: Based on the ACF and PACF plots, an ARIMA(2, 2, 2) model was chosen.
• Model Fitting and Forecasting: The ARIMA model was fit to the data, and forecasts for the next 24 months were generated.
• Visualization: The actual data, forecasted values, and confidence intervals were plotted to provide a comprehensive view of the projected CPI.

This approach provides a robust forecast by incorporating both autoregressive and moving average components, accounting for trends and seasonality in the core CPI data. The confidence intervals offer a range of possible future values, helping to understand the uncertainty associated with the predictions.

Full disclosure: long SPY calendar call spread, long TLT puts

The post Fed’s Powell Avoids Rate Cut Talk While April CPI Keeps Hope Alive appeared first on Inflation Watch.

The article includes some statistics on the huge difference in price trends on various food items in India:

“The cost of bananas in New Delhi is up 50 percent over the year, while paneer – a form of cottage cheese – has risen 26 percent to 145 rupees per kg.

Yet other food prices are falling. Staples such as tomatoes and potatoes, which peaked earlier in the year at levels that caused great stress to poorer families, have seen prices moderate in recent weeks.”

The post Asian inflation may finally be cooling appeared first on Inflation Watch.

First, Bernanke made it clear he thinks gold is not a good indicator of inflation expectations. Now, the Federal Reserve Bank of San Francisco has produced research that could convince the Fed to insulate itself from the inflation expectations of average Americans in “Household Inflation Expectations and the Price of Oil: It’s Déjà Vu All Over Again” by Bharat Trehan (thanks to Bill Fleckenstein for calling this article to my attention).

Household inflation expectations have risen to 4.5% from 3% at the end of 2010. Fortunately for the Federal Reserve, its empirical research seems to show that household expectations have become inaccurate and irrelevant for monetary policy:

“This Economic Letter argues that the jump in household inflation expectations is a reaction to the recent energy and food price shocks, following a pattern observed after the oil and commodity price shocks in 2008. The data reveal that households are unusually sensitive to changes in these prices and tend to respond by revising their inflation expectations by more than historical relationships warrant. Since commodity price shocks have occurred relatively often in recent years, this excessive sensitivity has meant that household inflation expectations have performed quite badly as forecasts of future inflation.”

Trehan admits that the University of Michigan’s Survey Research Center shows that households had been pretty good indicators of future inflation from the 1970s to 2000. However, over the past several years, the increased volatility in the prices of food and energy have misled consumers to anticipate more future inflation than is warranted given low levels of existing core inflation:

“The recent jump in the Thomson Reuters/University of Michigan measure of household inflation expectations appears to be related to increases in the prices of energy and food, similar to the jump observed in 2008. The size of this response to noncore inflation cannot be justified in terms of the historical relationships in the data. This disproportionate response is probably the reason why household inflation expectations have not done well as forecasts of future inflation in recent years, a period of volatile food and energy inflation. The poor forecasting performance argues against reacting strongly to the recent increases in household inflation expectations.”

Moreover, recent increases in inflation expectations are not justified by changes in monetary policy. Trehan speculates that…

“It’s also possible that households’ sensitivity to noncore inflation goes up following substantial, sharp increases in the price of energy and food items, such as those that occurred in the 1970s and over the past few years…This similarity to the 1970s is unsettling because it suggests that consumers are not accounting for the ways monetary policy has changed over this period.”

I assume this claim means that the Federal Reserve’s monetary policies have improved since the 1970s. In my opinion, we have seen even less reason to trust in the Federal Reserve’s policies to the extent that these policies “fix” economic problems in such a way to help set up the next crisis. These crises build while the Federal Reserve tends to reassure that it has everything under control and/or there is nothing happening to cause concern.

I would challenge the historical record and related regressions to suggest we need to consider whether the structural underpinnings of inflation are changing in ways that the Federal Reserve will be slow to recognize. Whether the Federal Reserve can do anything about these changes is another question.

The post The Federal Reserve May Choose to Ignore the Inflation Expectations of Households appeared first on Inflation Watch.

“The ongoing rise in China’s inflation rate is making headlines today, but U.S. inflation is not too far behind, as this chart shows. It’s not surprising that inflation should be moving higher both in China and the U.S., since China has essentially outsourced its monetary policy to the U.S. Federal Reserve by pegging the yuan to the dollar. Chinese inflation is somewhat more volatile than ours, and that is also not surprising since its economy is smaller and less burdened by long-term supply and labor contracts. If China has an inflation problem, then so does the U.S. It will just take longer for the problem to become obvious in the U.S.”

This piece is a must-read.

Disclosure: author is long TIPS

The post Looking at CPI from all angles suggests an imminent problem appeared first on Inflation Watch.

“The Swiss National Bank (SNB) is maintaining its expansionary monetary policy. It is leaving the target range for the three-month Libor rate unchanged at 0.0–0.75%, and intends to keep the Libor within the lower part of the target range at around 0.25%.”

The SNB goes on to outline economic conditions around the globe highlighting the strengths and weaknesses. Interestingly, the Swiss economy has done relatively well despite the record strength in its currency. Specific areas of weakness are tourism and exports, both highly dependent on relative exchange rates.

On balance, the SNB found it necessary to increase its near-term inflation forecast a bit. The SNB also again acknowledged that it cannot maintain its current expansionary monetary policy given its expectation that inflation will exceed its upper limit of 2% by 2013. With its currency too strong for comfort, the SNB is clearly trying to wait as long as possible before moving to head off these higher than desired inflation levels. It is a dilemma more and more central banks will feel as the U.S. dollar continues to sink against every major currency.

From the monetary policy statement:

The post Swiss National Bank increases near-term inflation forecast appeared first on Inflation Watch.

“A package of Oscar Mayer cold cuts. A pair of Nine West boots. A Whirlpool washing machine.

By the fall, people will most likely be paying more for each of them, as rising prices hit most consumer goods, say retailers, food companies and manufacturers of consumer products…

…Many big companies, including Kraft, Polo Ralph Lauren and Hanes, say they cannot hold off any longer and must raise prices to protect some profits.”

While such talk has come later in the economic cycle than we expected, its arrival should be taken seriously. The NYT article quotes analysts who take the other side of the story, for example claiming that consumers will not pay the higher prices. However, with corporate profits at historic levels despite extremely high unemployment, we should not underestimate the inflationary pressures that could stick once companies feel compelled to finally pass on their higher costs to consumers.

The post Inflation scheduling an autumn arrival in the U.S.? appeared first on Inflation Watch.

Apparently, there is at least one other person forecasting a time when inflation is “everywhere.” CNBC interviewed Arun Motianey, director of fixed income strategy at Roubini’s RBG Capital and a protege of economic professor Nouriel Roubini:

“We’re heading into a world of inflation because we are highly indebted and we are indebted here in the US economy in the household sector and in the financial sector…It’s going to be inflation everywhere and it’s going to happen really through the weakness of the US dollar…Then inflation in those other parts of the world that are expecting appreciating currencies, they’re going to inflate as well because that’s the way you ultimately correct this [economic slowdown].”

Motianey asserts that this inflation is required to avoid the apparently worse alternative of deflation.

The post Roubini protege expects inflation all around the world appeared first on Inflation Watch.