June's Inflation Report Took a July Hike Off the Table. The Warsh Fed Is Not Done.

What the June Report Actually Said

The Consumer Price Index the government's main gauge of what a basket of everyday goods and services costs fell 0.4 percent in June. That brought the headline inflation rate down to 3.5 percent from 4.2 percent the biggest monthly improvement in more than a year

Eliminate food and energy and you will get Basic CPI the number the Fed actually follows because it is less noisy. The core was stable month over month and declined to 2.6 percent year over year

Wholesale prices told the same story. The Producer Price Index which measures what companies charge each other before products reach stores fell 0.3 percent and the core PPI rose just 0.2 percent. When both consumer and producer calm down at the same time it is usually a real sign and not a one-month fluke

At first glance this is about as clear a disinflation footprint as a central banker could ask for. Look at which line did the job and it gets more complicated

Why the Fed Almost Hiked Anyway

Kevin Warsh took over as chairman of the Federal Reserve in May and arrived with a reputation as a hawk someone who is more worried about inflation breaking out than growth slowing. Heading into July futures markets had priced in a real possibility that his first meeting would generate another quarter-point increase to show that the new committee meant business

The June data made it a tough sell. You don't tighten policy to the best inflation in a year without a compelling reason and the labor market hasn't crumbled in a way that would force the issue. So traders quickly abandoned their bets on a move in July

The interesting thing is what they didn't abandon

A Worked Example: Backing Out How Much Was Energy

You can do this on paper in about a minute and it changes the way the report is read

The CPI is a weighted average. The heading is the basic plus food and energy that the core excludes each weighted by its share of household spending. In round numbers basic foods make up about 79 percent of the basket and food plus energy makes up about 21 percent. Those weights change a little each year so treat them as illustrative rather than exact

Now write what we know. The stock fell 0.4 percent on the month. The core was flat so its contribution was zero. Plug both into the weighted average:

0.79 times 0 plus 0.21 times x equals negative 0.4

Solve for x and you get about negative 1.9 percent. In other words the food and energy portion of the basket had to fall on the order of two percent in a single month to produce an overall drop of four tenths while everything else held still. Food doesn't move like that. Energy does

Then run it the other way around which is the part you should be worried about. Keep the core flat again and assume that the energy complex simply retraces that same two percent next month. The arithmetic is symmetrical. The headline rises 0.4 percent and the year-over-year rate that just posted 3.5 percent rises back to the three highs

Nothing about the underlying inflation picture would have changed. Core would have done the same thing in both months. All the back-and-forth from the headline the one who took a rate hike off the table and put it back on would have come from a single volatile factor that no central bank controls

The Energy Problem Hiding in the Data

Here's the problem that an attentive reader sees. A large part of June's improvement was due to the decline in energy prices which fell after a ceasefire in the Iran conflict briefly calmed oil markets. That ceasefire has since broken down. Tensions around the Strait of Hormuz the narrow shipping route through which much of the world's maritime oil is transported have flared again and the national average price of gas has risen back to $4 a year.gallon

Energy-driven disinflation is the least durable. If crude oil continues to rise through the summer the same force that flattered the June report will reverse it and the headline 3.5 percent could rise again. That's exactly the scenario that keeps a hawkish president up at night

It's also the scenario that a hawkish president is least prepared to fix. Raising the federal funds rate has no impact on the number of oil tankers that can pass through a strait. Monetary policy works on demand and a supply shock raises prices while reducing production pushing the Fed's two goals in opposite directions at the same time

Case Study: 2008 and the Central Bank That Chased the Headline

There is a well-documented example of what happens when a central bank treats a power spike as an inflation problem and it's worth looking into because the setup rhymes with it

During the first half of 2008 oil reached approximately $147 per barrel the highest price ever recorded up to that point. Headline inflation followed. The US headline CPI peaked at around 5.6 percent year-on-year in July

In July 2008 the European Central Bank raised its main policy rate to 4.25 percent explicitly to defend its inflation target against a headline rate that had surpassed 4 percent. By the announced logic it was defensible. Inflation was well above target and the bank had a mandate

The timing turned out to be about the worst possible. The financial system was weeks away from shutting down oil plummeted from $147 to less than $40 in a matter of months and headline inflation went from an emergency to outright deflation. The U.S. headline CPI was around negative 2 percent in mid-2009. The ECB was cutting aggressively in October having adjusted to the brink of the deepest recession since the 1930s

The instructive part is that core inflation never told that story. Core remained dull the entire time. The bank that saw the headline was taken aback and the signal that would have kept it calm was in the same statement

So when traders abandoned their rate-hike bets in July following a report whose improvement came from the energy sector they were not complacent. They refused to repeat a specific and costly mistake

What the Bond Market Is Telling You

Treasury yields the interest rates the government pays to borrow fell on the news. The 10-year bond closed near 4.525 percent and the 2-year bond near 4.124 percent each down a few. base points (A basis point is one-hundredth of a percentage point). Lower yields after weak inflation mean bond investors believe the Fed is closer to ending than restarting hikes

The 2-year bond is below the 10-year so the yield curve is no longer as inverted as it was during the tightening cycle. That normalization is what a soft landing in the bond market looks like a slow return to a world where longer-term loans pay more than short-term ones

The 2-year bond is the clearest instrument to look at here because it is essentially the market's average estimate of the policy rate over the next two years. It moved a few basis points. This is not a market that prices the end of an inflation cycle. This is a market that pushes its odds and waits

Where the Rule of Looking Through Energy Breaks

I just spent several sections arguing that the core is the honest number so let me argue the other side because the case against it is real and has been real before

The first crack is second-round effects. Energy isn't just a commodity it's an input. Diesel prices fuel freight costs freight costs fuel the price of everything trucked and utility bills fuel the cost of running a factory or a restaurant. If oil stays high long enough it stops being a food and energy item and starts showing up inside the core. At that point looking through it means looking throughof what is causing the problem

The second crack is expectations and this is the one that killed careers in the 1970s. Households do not experience underlying inflation. They experience the pump and the supermarket bill. If visible prices continue to rise people demand bigger increases regardless of what the base index says and wage growth turns a supply shock into long-lasting inflation. Central bankers who dismissed oil shocks as transitory supply events found that the public had turned them into everyone's expectations.modes

The third crack is more technical. In reality the Fed does not target the CPI. Its target is PCE inflation which weights the components differently in particular giving much less importance to housing. A CPI figure may flatter or scare you relative to the figure for which the committee is formally responsible

My conclusion is not that the core wins. It's that the core is the correct default and a bad religion. Look through the energy while the motion is back and forth. Stop looking at it the moment it stops spinning

How I Actually Read a CPI Print

My routine on the day of my release is brief and deliberately unglamorous

I read the main content month after month first before anything else and preferably before I see a headline anywhere. The year-over-year figure is a twelve-month average that tells me mainly about last fall. The monthly core number is the closest thing to a news item in the release

Second I do the subtraction from the worked example above. Headline minus core weighted tells me how much of the story a barrel of oil is. If the answer is most I rule out the market reaction that follows in both directions

Third I check shelter separately because it is the largest single core component and is measured with a long lag. Housing can keep the core index high for months after market rents have already stopped rising making the core index appear more unstable than the economy actually is

Fourth and this is the habit that has helped me the most I write what would change my mind before reading someone else's opinion. For this report mine is simple. Two consecutive underlying numbers at or below 0.2 percent per month with stable energy and I think the fight is really won. One more energy-driven headline change and I learn nothing

My read right now is that Warsh is not lying and that the market is correctly pricing in a pause rather than an end. I'd rather be early in calling the job unfinished than smart in calling the top. This is opinion not fact and certainly not investment advice

Reading October

Even after discounting July markets still assign a roughly 60 percent chance that rates will end the year between a quarter and a half point higher likely at the October meeting. In other words investors think Warsh is pausing not stopping

The Fed has traded a July hike for optionality the freedom to act later if energy prices undo the progress it just made. It's a really good trade for a new chair. A July hike would have spent credibility on a data point that can be reversed. Waiting costs almost nothing if inflation continues to cool and costs a late meeting if it doesn't

For a student trying to read the macroeconomic picture the lesson is that the Federal Reserve reacts to trends not headlines. One report changes the odds. It takes a series of them to change destiny

The Bottom Line

The June inflation report was actually good good enough to rule out a July rate hike and allow the Warsh Fed to breathe. But the improvement was largely based on cheaper energy and the arithmetic shows just how much. With the core flat on the month the food and energy share had to fall about two percent to produce that headline and the same two percent can immediately return

The 2008 episode is the reason that matters. Headline inflation peaked at around 5.6 percent with oil at $147 core inflation never moved the ECB rose to that level and within months the problem had completely reversed. The way to defeat central banks is to chase a volatile input

The market message is measured. The fight against inflation is being won slowly and it is not over. Let's look at oil not just the CPI headline to guess what the Fed will do next and look at underlying oil for the moment when oil stops being a separate item and starts seeping into everything else