Oil spiked higher this week, regaining $90 a barrel as the dollar rose and U.S. Treasury bonds sold off. It is not hard to guess what the stock market did as a result. 

 

Down.

 

I'm sure you are just as tired as I am watching this Kabuki show unfold in the Middle East. Not only is the Straits of Hormuz shut in, but now Iran's proxy, the Houthis, have opened a second front along the Red Sea. They have already attacked two Saudi oil tankers and have vowed to close that oil avenue off from any further shipment of oil from Saudi Arabia.

 

Brent crude is up more than 40 percent in three weeks. Readers only need to fill up at the pump this week to realize the cost of this ongoing travesty. The Ten-year U.S. Treasury bond is now yielding 4.67 percent. Traders are dumping U.S. Treasuries as well as stocks in anticipation that by August the inflation data will be rebounding substantially. That would put an interest rate hike back on the table by September by the Federal Reserve Bank.

 

It is a tangled web this administration has woven. There were reports on Friday that Pakistan, with support from China, was seeking to revive negotiations between the two adversaries. That dropped oil prices by 5 percent to around $88 a barrel. Hope springs eternal I guess when  dealing with this war.

 

Oh, in case I forget, the president has just slapped a whole host of new tariffs (10-12 percent) on world trade, manufacturing a new excuse (forced labor in 80 countries) as justification. This adds yet another layer of price increases consumers will be receiving in the months ahead since we now know you and I are paying most of these tariff costs.

 

All the goals of this administration's economic policies, touted by U.S. Treasury Secretary Scott Bessent — reduce budget deficits, boost growth and increase energy production — have remained pipe dreams. Instead, interest rates are reaching new highs, spending and deficits are off the charts, and oil, rather than declining, is skyrocketing.

 

The AI trade has faltered as well. The recurring worry that the large mega-cap tech companies are spending too much money plagues the markets. The fate of this area hinges on the outlook for 2027 capital expenditures growth from the hyperscalers like Google, Meta, Amazon, and Microsoft. Currently, Wall Street analysts are expecting capex to grow by 28 percent next year. That's up from 23 percent two weeks ago and before Google's second quarter earnings announcement on Wednesday night.

 

Google once again raised its estimate of how much more it is planning to spend on AI this year, from $190 billion to a range of $195 billion to $205 billion. The stock cratered on the news despite a blockbuster revenue growth of $119.8 billion, up 24 percent from a year earlier. And what Google is doing, its competitors will do too. By the end of the earnings period, we could see that number increase to 37 percent.

 

Given that all these companies have whittled down their cash due to this monumental spending, investors expect that the only way to increase spending further will be for these companies to sell more stock and raise debt, thereby diluting existing holders. Even if they succeed, there is still no guarantee anytime soon that these companies will see the kind of payoff that is necessary to the bottom line given the amount of money involved.

 

Tesla was another dud. Auto sales are falling, and capex in all his tomorrow ventures, including AI, is exploding higher. Combined with the 50 percent decline in the price of SpaceX, Elon Musk is keeping a low profile lately.

 

Regular readers know that I entered the July-August period rather cautiously. I believe we are in a normal mid-summer consolidation in a mid-term election year. It appears as if Donald Trump is working overtime to ensure Democrats win that contest. That adds even more uncertainty to an equation already burdened by the possibility of a "massive attack" in Iran and therefore further spikes in energy prices. So far, second-quarter earnings have been on target for the most part. Next week, Microsoft, Meta, Apple and Amazon report on Wednesday and Thursday. Their announcements will largely dictate which way technology goes in the short term.

 

I see two possible outcomes for the markets over the remainder of the summer. Both would bring with them high volatility. The first is that we chop around here. Since technology is leading this pullback, I see it trading in a range of a little above and below 680 as reflected in the main Technology ETF (QQQ).

 

The S&P 500 Index, supported by the rotation I have discussed previously (that is out of tech and into sectors like healthcare, utilities, industrials, etc.), could remain at this 7,400 level. Once again, you could still see 50-point swings weekly or even daily in the index.

 

If, on the other hand, Trump allows his emotions to play out with few in the White House willing to talk him down, a sudden escalation in the war might occur. Oil prices spike much higher as a result. In that case, the first stop on the S&P would be 7,300 (testing a double bottom). If that fails to hold, we are looking at 7,200 (cycle lows). Technology would have an even bigger decline.

 

In an era of populism, distrust in government is rampant. That lack of trust has seeped down into just about everything one reads or hears. Government data is no exception.

 

As I wrote in my last column, our own president fired the head of the Bureau of Labor Statistics, claiming the BLS was "broken." He further claimed it would need new leadership "to restore trust and confidence in the BLS's data."

 

Since then, although there has been an acting BLS commissioner, no permanent replacement has made it past the Senate. Six months ago, Trump picked a veteran economist, Brett Matsumoto, as his nominee. This week, a Senate committee voted (12-11) to advance his nomination, but it is unclear whether the full Senate will vote to confirm the nomination. 

 

In his confirmation hearing, Matsumoto recognized the potential for further manipulation and politicization of the BLS. "It is important for the public to be confident that decisions at the BLS are being driven by science rather than politics," he said.

 

If it is science he is after, he may first need to hire qualified subordinates. Fully one-third of the top leadership positions at the BLS are vacant. That may be difficult, since there is a hiring freeze in effect, plus he must deal with a swath of deferred resignations and early retirements. That could be an uphill battle given that the Supreme Court gave the president carte blanche to fire independent government regulators despite federal job protections. As for the money needed to upgrade the governmental data systems, that too will be problematic.

 

In the meantime, meddling with government data, at least on the inflation front, continues. Last month, the Bureau of Economic Analysis, which calculates the Personal Consumption Expenditures Price Index (PCE), announced changes to how they plan to track data. The new method for capturing data and calculating price changes across three subcategories will be revised. In essence, the changes will make the numbers look better (with less inflation) than in the past.

 

Analysts estimate it will reduce core PCE inflation by about 0.2 percentage points. You may remember that Stephen Miran, a Trump advisor and chief architect of "reciprocal tariffs," was appointed to the Fed for a six-month stint and then replaced by the new Fed Chairman Kevin Warsh, another Trump appointee.

 

Miran, along with two Fed staff economists, is behind this effort to alter the PCE, the Fed's main inflation index used to determine the nation's inflation rate. I am sure the Fed will have logical, technical reasons to justify this improvement in their key inflation indicator.

 

At the same time, Fed Chairman Warsh is creating five new policy review task forces to investigate communications, balance sheet policy, productivity and jobs, inflation frameworks, and data. It seems more tinkering is ahead of us.

 

I guess it is pure coincidence that these changes come at a time when mid-term elections are a few months away. We won't know the result or the changes (if any) that may occur until next year. I'm hoping it helps rebuild trust and accuracy rather than the opposite.

 

In my career, I have seen instances of politically motivated meddling in places like Greece, China, and Argentina, to name just a few. In every case, investors lost faith in the data of the country in question, leading to higher borrowing costs.

 

We already have an almost daily problem with nonpublic information that can move markets coming from both within and without the government. Some of it is legal (if questionable); some of it involves leaked advance information from government sources. It doesn't seem that any of the regulatory agencies is willing or capable of stopping it.

 

Taken together, the leaks and inaccuracies among government agencies, both real and intended, are contributing to a deepening sense of distrust and cynicism among voters. Further neglect and delay in regaining that gold standard of government statistics that the country earned over decades should not be taken lightly.

 

Misery loves company, or so they say. There was plenty of that this week as declining technology stocks dragged down the indexes.

 

While the stock market had been stuck in a rut most of the month, the tone changed this week. Markets felt the pain as the leading AI -fueled semiconductor index was walloped. However, appearances can be deceiving. There were some pockets of safety that did well as rotation strategies replaced momentum plays among traders.

 

Quarterly earnings have begun. The estimates call for a 23-24 percent gain in earnings this season. That follows a 26 percent+ record last quarter. Bank stocks were the first to report, and most did not disappoint. Goldman Sachs' profit, for example, was up 78 percent from the prior year while JP Morgan soared 41 percent. It may come as no surprise that investment banking fees had a lot to do with those results.

 

The large-cap banking index made new highs for the year even as AI technology and the semiconductors index suffered additional selling. Boring old utilities, consumer staples, and health care outperformed as most of the AI darlings languished.

 

SpaceX, the Musk deal-of-the-century IPO that I warned readers not to chase, is now down to $124/share, well below the $135/share IPO price. Last week's broker-hyped offering, the "must have" Korean-based ADR, SK Hynix, has also been a dud (minus-37 percent).

 

Not all technology has done poorly. After being ignored or sold down for weeks, the Magnificent Seven stocks have added $1.5 trillion in market value in July, while semiconductor stocks, excluding Nvidia, have erased nearly $1.7 trillion in market value. Software companies, another casualty of AI predominance, have come back from the dead. Forty-four out of 51 software stocks in the Yahoo Finance industry basket are up for July with a median gain of 6 percent.

 

Given the high valuations of most stocks, it seems investors are quick to punish and just as quick to reward. Those companies that disappoint, failing to live up to investors' expectations, are quickly taken to the woodshed. IBM announced weak preliminary results, and the stock fell 25 percent, the largest decline since at least 1968. Netflix also disappointed and opened down 10 percent on Friday.

 

The next two weeks should be interesting as more companies report. I warned readers in weeks past that, this time around, quarterly earnings will see investors take a much more selective approach to companies based on their results and guidance. Evidence so far indicates I am not far off the mark.

 

This week, we also had the results for both the Consumer Price and Producer Price Indexes for last month. As I predicted, both numbers fell well below street expectations. I also expect next month's numbers to be weak as well (unless Trump's Forever War pushes oil prices higher still). Markets pushed higher for a day in celebration, but it didn't last long.

 

Kevin Warsh, in his first appearance as Federal Reserve Chairman before the House Financial Services Committee, threw cold water on the monthly inflation numbers. He pointed out that one or two data points do not make a trend.

 

Warsh said, "The longer prices have been above the inflation target, it's usually a bit harder to dislodge them and get them lower. Our job, my commitment to you, is to take sticky prices and to unstick them." That may be music to the ears of Main Street (and me) but do nothing for the financial market's hopes of easier monetary policy this year.

 

The bullish tone of the markets preceding and just after the Fourth of July has come and gone. As readers know, I have approached July and August with caution. Since the holiday, investor sentiment and fund flows have waned, while the technology sector has come under more pressure.

 

This week, we saw further evidence of that as the Nasdaq declined more than 3 percent, the S&P 500 dropped 1.43 percent, and the Russell small-cap index, the best of the bunch, maintained its bullish posture. It declined by less than half a percent.

 

It is no surprise to see the areas that went up the most experience the most severe declines. It is how markets work. Profit-taking in semiconductors was in full force this week. While I do expect bounces along the way, I think over the next few weeks we will see further downside.

 

The data system supporting U.S. economic decisions is losing staff and funding. What's worse is the diminished trust the financial markets have placed in that data.

 

There are roughly 13 primary statistical agencies embedded within various cabinet departments and federal agencies. There are another 100 or so small decentralized statistical offices scattered across different agencies. A major problem all these agencies face is in data collection. The process is antiquated, and the system overall is worse.

 

It is a system badly in need of modernization, innovation, and a huge technological overhaul. The introduction of artificial intelligence might help. To do so, we need to bring in outside partners in the private sector. The government needs to identify state-of-the-art players in AI but also in other areas. Those that have the ability and experience in data gathering necessary to come up with new methods and strategies to obtain data in the most efficient and accurate ways possible.

 

Currently, in this age of cost-cutting through bureaucratic reductions, there is little enthusiasm in Congress or the White House to address the problem. Few data advocates exist in today's mostly populist Congress. Since bureaucrats rarely, if ever, get to talk to legislators, it's an out-of-sight, out-of-mind kind of situation.

 

As things continue to fester, another issue has also seeped into the equation. A growing belief among many is that this data is trustworthy. Is this data truly unbiased, or has it become politicized?

 

Readers may recall that a year ago, President Trump fired the head of the Bureau of Labor Statistics, Erika McEntarfer. Trump, fresh off his second presidential win, was incensed when the BLS released a revised employment statistic. The BLS reduced the number of jobs thought to have been created from March 2024 to March 2025 by 91,000.

 

Those 91,000 ghost jobs rank among the largest revisions in recent decades. "Biden's economy was a disaster, and the BLS is broken," Trump claimed, "This is exactly why we need new leadership to restore trust and confidence in the BLS's data on behalf of the financial markets, businesses, policy makers and families that rely on this data to make major decisions."

 

Trump had a point. Ever the political animal fresh off a tight race, the president recognizes the iron law of American politics — economic conditions drive election results. Historically, good economic indicators (such as low unemployment, rising wages, and low inflation) often help incumbents, while poor performance hurts them.

 

Conspiracy theorists immediately accused the former administration of padding the numbers in hopes of tilting the voters toward the Democrats' candidates. Whether that was true or not, experts insist that the process that produced that number, while imperfect, was transparent and has been used by the agency for decades. This conclusion was largely echoed by the Chairman of the Federal Reserve.

 

Back in December 2025, former Fed Chairman Jerome Powell explained that the BLS has consistently overstated jobs by at least 60,000 per month since April of 2025. The culprit was a monthly BLS estimate of how the labor market is affected by business openings and closings. The estimate, known as the birth-death model, guesses at the jobs gained and lost due to closings. This, Powell said, was "something of a systematic overcount" that would likely see big revisions to job growth numbers.

 

Nonetheless, Trump, without any evidence, fired the BLS chief, claiming the data was being rigged. That move had ramifications that reverberate today. New research by four economists at the Center for Economic and Policy Research indicates that because of Trump's claims, there was a sharp increase in policy uncertainty among American businesses. That reduction in trust and increased uncertainty depressed economic activity as corporations pulled back on investment. 

 

The hit to the economy was in the vicinity of $20 billion. Next week, I will discuss additional ramifications of doing nothing to change the status quo of how government data is gathered and distributed. 

 

The above risks can be a lethal brew for investors. And yet, markets barely budged by the end of the week. Quarterly earnings expectations are buoying the averages, with 23 percent earnings growth expected to vindicate the bulls.

 

Markets have moved beyond Trump's War despite this week's conflict. They are focusing instead on the upcoming inflation data, the start of quarterly earnings next week, and another Fed meeting. In the meantime, technology still struggles.

 

Volatile moves have sent the Nasdaq and the semiconductor index plummeting only to see them spike higher on a day-by-day basis. Traders call it a calm session if these stocks close with less than a 1 percent move each day. Last month's mega IPO is a case in point.

 

Readers recall all the hype over the SpaceX IPO. Priced at $135 a share, opened at $150 and skyrocketed to $212. Friday it traded at $147-$148. Not pretty, if you chased it. Today, we have another one.

 

This time, it's the listing of a South Korean memory chip leader, SK Hynix (SKHYV). It is the largest American Depository Receipt (ADR) offering ever ($26.5 billion) and one of the largest equity offerings in history. It is the global leader in High-Bandwidth Memory (HBM). Why is their product so important? Because without HBM, there would be no AI.

 

At $149 per ADR, it will be equal to 1/10th of a South Korean share. With a market value of more than $1 trillion, it is the second-most valuable company in South Korea. The media claims the offering is seven times oversubscribed (versus SpaceX's five times). And like SpaceX before them, chasers ‘gotta get some.'

 

No, never mind that the memory stock has garnered a sevenfold increase in its stock price over the past year. If it performs the way Elon Musk's SpaceX ("to the moon and beyond") did, we could see another price spike before traders cash in. At around midday, the ADR was up $17 percent from its listing price. And while the financial media focuses on this offering, it wasn't the only event of the week.

 

The president and his forever war kept investors on their toes. He now says the ceasefire that never was is over, but the talks will continue. How bombing more Iranian military targets is going to do anything to change the status quo is beyond me. As this week's NATO conference has shown, despite Trump's bravado, most nations still need to flatter, or at least humor him, if they want to remain under the U.S. military umbrella. Strategically, they need to maintain that relationship, at least in the short term.

 

Fortunately, the markets have moved on. The rotational trends in the markets have helped keep the main averages steady most of the time, despite Trump's social media posts and comments. The oil price has risen slightly (over $71.80 a barrel) from $67, but the technical trend still points to further downside.

 

Bond yields have risen to the top of their range with the U.S. 10-year Treasury bond hitting a high of 4.57 percent this week. As you might imagine, Trump's military strikes and the subsequent short-term rise in oil prices immediately had traders rushing for the exits in some areas and chasing stocks on what had been the ‘war trade'.

 

Here's how it works. The narrative is quite simple, really: missiles fly, oil prices spike, inflation expectations rise, and so the story changes to the Fed having to raise rates. That's it in a nutshell. The opposite occurs whenever the narrative shifts toward peace, the opening of the Straits of Hormuz, etc.

 

Next week will be critical for the bulls. We get another Consumer Price Index reading. The Street is expecting cooler inflation numbers for June. I agree. I expect weaker numbers in July as well. That should be good for the markets.

 

Stock prices have already been bid up in anticipation of good earnings. If management's ‘beat' and talk up future guidance on sales and earnings, then all is well with the world. The rally continues as the indexes grind higher. We all know what happens if companies fail to live up to expectations. We may see investors become a little more selective. The AI trade may shift from buying "everything AI-related" to buying stocks worth holding, rather than those that are not.