The Markets Versus A Hawkish Fed (Technically Speaking For The Week Of 3/21-3/25)

Bull and Bear Symbol with Stock Market Concept.

undefined undefined/iStock via Getty Images

Let’s review this week’s economic releases.

The Chicago Fed’s National Activity Index points towards above-trend growth:

The Chicago Fed National Activity Index (CFNAI) moved down to +0.51 in February from +0.59 in January. Three of the four broad categories of indicators used to construct the index made positive contributions in February, but two categories deteriorated from January. The index’s three-month moving average, CFNAI-MA3, ticked down to +0.35 in February from +0.37 in January.

Personal consumption spending contributed the only negative data:

The contribution of the personal consumption and housing category to the CFNAI fell to –0.04 in February from +0.21 in January.

Building permits were modestly lower:

Building permit data

Building permit data (Census)

Here’s a chart of the data:

1-unit building permits

1-unit building permits (FRED)

1-unit permits started to rise last September. The latest dip is likely a modest cooling during a period of rising activity.

We now have a complete set of new housing industry data. Existing home sales were lower:

Total existing-home sales,1 Existing-Home Sales, completed transactions that include single-family homes, townhomes, condominiums and co-ops, sank 7.2% from January to a seasonally adjusted annual rate of 6.02 million in February. Year-over-year, sales decreased 2.4% (6.17 million in February 2021).

Here’s a chart of the data:

Existing home sales

Existing home sales (Calculated Risk)

This data has been consistent for about a year.

New home sales were off:

New home sales

New home sales (Census)

Here’s the chart:

New home sales

New home sales (FRED)

The recent declines are likely due to rising rates. But trends (higher rates and lower sales) should continue throughout the year.

Durable goods orders were lower:

Durable goods orders

Durable goods orders (Census)

Here’s the data:

Durable goods orders and durable goods ex-transportation

Durable goods orders and durable goods ex-transportation (FRED)

This is the first dip in total new orders in nearly two years. Total orders ex-transportation, however, are still rising.

At worst, the above data is neutral. But a more accurate description is that it’s softly positive.

Now let’s turn to the charts, starting here:

1-year SPY, QQQ, DIA, and IWM

1-year SPY, QQQ, DIA, and IWM (StockCharts)

All of the indexes have broken key long-term support. That means we’re now in a new market phase.

6-month SPY, QQQ, DIA, and IWM

6-month SPY, QQQ, DIA, and IWM (StockCharts)

But the shorter-term charts show that the markets have mostly broken the recently-established downtrend, with the exception of the IWM.

This means we need to assess the possibility of a new upward trend emerging.

Technical Indicators

Technical Indicators (Author’s Data)

The above spreadsheet is from the author’s Stockcharts.com data. It is based on the exponential moving averages from 1-year charts. On the bearish side, the longer-term indicators are still negative. All 20-day EMAs are below 50-day EMAs, two 50-day EMAs are below 200-day EMAs, and all 200-day EMAs are either neutral (moving sideways) or bearish (heading lower).

But shorter-term data is positive. All the 10-day EMAs are above the 20-day EMAs, and all 10, 20, and 50-day EMAs are moving higher.

In other words, the markets want to turn more bullish but that hasn’t translated into long-term bullishness yet.

The biggest challenge to a prolonged rally is the Fed. As I noted earlier this week, the Fed has clearly shifted to a hawkish stance. Three presidents have stated they are in favor of a 50-basis point hike to stem inflationary pressures.

And that leads to the final point: the old trader’s adage of, “Don’t fight the Fed.” Higher rates are typically bearish for stocks. That greatly diminishes the possibility of a longer-term rally.

Be the first to comment

Leave a Reply

Your email address will not be published.


*