Capital Flows and Asset Markets

Generally speaking the fund management community thinks of the dot com boom and the housing bull markets as “bubbles”. Looking at US net worth to GDP, the dot com bubble say net worth to GDP rise form 375% to 450% from 1995 to 2000. And the housing bubble saw it rise from 400% to 475% of GDP. This current cycle has seen “the bubble” rise from 400% to 600% and is till at 560%. With US equities back at all time highs, hard to argue that “the bubble” has burst.

The top of the dot com bubble saw gold outperform the Nasdaq. Something that started to happen this year. If you were thinking in terms of a “tech bubble” there have been plenty of reasons to think we were at the top. When Chinese tech stocks got regulated to low growth in 2021, I assumed a similar outcome awaited US stocks. This has not happened.

When Deepseek came out with a low cost AI agent, I thought the whole AI stack would need to reprice lower. Nvidia weakened for a while, but has now touched all time highs again.

When I read about Mark Zuckerberg doing a USD14bn aquihire I think we must be near a top of a bubble. But I thought similar things when he bought WhatsApp for USD 18bn - which now looks like a bargain.

What I think all of this has in common is the ever widening power of the Silicon Valley idea of blitzscaling. While Amazon is a powerful example of the power of growing rapidly first, and profits later, from an investing perspective, Softbank is another great example. Masayoshi Son only wanted companies to grow rapidly. The 99% market value lost from 2000 to 2002 did not dissuade him from this style.

The question I am asking myself is that every corporate and every investor now understands the blitzscaling methods. Palantir on 107 EV/Sales is not a bubble, but the huge losses it racked up, and the building of product that can rapidly rolled out to the US government is intensely valuable. There is also a reflexive nature to the blitzscaling model. The company with the largest market capitalisation is most likely to be able to invest the most to capture the most market share in the future.

This is where the Blitzscaling ideology gets hard for me. When I look at the cloud/data centre area, I can find plenty of stocks indicating that there is a potential overbuild coming. Equinix is a data centre REIT - and has highlighted slower growth in the short term. Stock is down 20% this year.

But on the other hand, newly listed CoreWeave which is building even more data centres has quadrupled from its listing in March. Which one is correct? Or has the market decided they want CoreWeave to be the blitzscaler here and realise hyper growth using its shares as a currency for growth? Its potential acquisition of Core Scientific would underscore this idea.

What I am trying to say, is that talk of a bubble is probably wrong. The huge dislocation between “growth” and “value” stocks reflects corporates and investors pricing in the success of blitzscaling. The trend change in growth versus value investing has been profound. And betting against this has been largely career ending.

One of the features of the blitzscaler model is that the amounts being invested into new areas has soared. Mark Zuckerberg is throwing large amounts of money at catching up in AI. Google which is very exposed to a potential shift to Chat GPT and other AI systems is spending even more. As far as I can tell Open AI does not make money, and given the appearance of DeepSeek, is not likely to in the short to medium term. But what we can see with Nvidia share price, this has not stopped ever increasing levels of Capex. As Amazon or Tesla has showed investors - you need to be with the growing and investing company no matter what. So where does it end? Well in China it ended with government regulations to reduce the pricing and market power of the leading tech companies. In many ways this is similar to the US government acting against Microsoft during the first dot com boom.

I used to think increased competition and falling prices would end this investing model. But on that basis I would need to be short Tesla, which is still priced as growth stock, even as sales growth has slowed and margins have fallen, and Chinese competitors like NEO have been destroyed. Tesla has not been a short.

Where does that leave me? Well away from the “blitzscaling” companies - markets have behaved much as I thought they would. Higher interest rates have acted as a weight on valuations and earnings. Blitzscalers however continue on their merry way. Spotify certainly now trades as the dominant stock in its area.

Companies like Uber and DoorDash, that seemed likely to suffer margin pressure from higher wage costs have both soared this year, even as near competitors (Lyft and Delivery Hero) have languished. A sign that the blitzscaling model has triumphed in this area as well. Blitzscaling is a problem for my style of investing - as I tended to like an industry, rather than trying to pick one stock. I would take weakness in Nio, Lyft, Delivery Hero as signs of problems in the industry - when in fact, it was a signal to buy the leader. The logic of this is that perhaps the Nasdaq is now the risk free asset - as the component stocks are now economically and politically untouchable blitzscalers. I launched Brumby Capital because I thought a pro-labour move in politics would be inflationary, which would put pressure on interest rates to move higher, and this would put pressure on equities. The beginning of 2025 saw a strong move higher in gold/nasdaq - but as of today, it has probably reversed half the move, and near the 200 MDA.

The problem with investing in the blitzscalers here is entirely political as I can not find a problem with their business models. The create a lot of share holder value, but relatively little taxation revenue. This continues to reduce the tax base, while adding to wealth inequality. China acted against this, and there is little question that the Chinese Community Party is in charge. There generally also seems to be far more competition in the Chinese tech space - which is leading to far more competitively priced tech. More competitive tech space, and lower prices has seen a huge gap open up between US and Chinese bond yields. At what point does the concentration of wealth in blitzscalers become a political problem? It seems inevitable to me - but timing is a key issue.

Blitzscaling being inflationary makes sense to me. The blitzscaler model is deflationary to start with, but once dominance is established, prices rise radically, and effective competition is virtually eradicated. For me, this is a “MARKET” problem if gold is outperforming stocks - and not a problem when stocks are outperforming gold. Given the bounce back in stocks - we stand again at a precipice. 2025 looked like a top to me - but perhaps we need an even more extreme move. On current trends, we should know which way the political wind is blowing soon.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.russell-clark.com/subscribe

It has been axiomatic that problems in the middle east would cause problems for the world, mainly through spikes in the oil price. Stagflationary 1970s was driven by the two oil price shocks, driven mainly by OPEC embargo on selling oil to the US. Over this period oil rose from USD 1.70 a barrel to USD35 a barrel driving rampant inflation. Believe it or not, this is before my time.

I am old enough to remember when Iraq invaded Kuwait. Oil prices spiked from USD 20 a barrel to USD 40 a barrel, and led to a brutal recession in Australia.

Unemployment in Australia surged to 11%, the highest level since the great depression. Getting a job was very hard, and remained so for quite some time.

So there is a muscle memory that energy price spikes are diabolical. Certainty markets were poor during the Russia induced spike in European natural gas prices.

But even the most heavily dependent nation on Russian gas, Germany, never saw a meaningful jump in unemployment.

So first of all, plainly energy prices do not flow into unemployment as readily as they used to. The rise of service economy versus a manufacturing economy is probably the main explanation. But there is a political explanation too. During the 1970s, OPEC AND the entire Soviet Block including China were net exporters of oil. Given that the communist block was not a market based economy, as long as they had oil supply economic growth would be fine - it was only market pricing economies that would suffer - hence OPEC could threaten the US with oil price surge, while not hurting its ally, the Soviet Union. China at the time was also an exporter. But obviously this is not true anymore.

China is now much more reliant on oil imports than the US. Does it make sense for Iran to hurt the one economy that has the military might to worry the US? Not really. Even more problematically for Iran is that Saudi’s have shown a willingness to raise production. They probably have another 2m barrels a day they could bring on line - given they are producing 9m barrels a day versus a peak of 11m barrels.

With Iran at about 3.3m barrels a day, even if that was shut down, the world could probably cope if the Saudis played ball - and there is every reason to think they would. There is also the question of does Iran even have enough money to live with out oil revenue for a prolonged period of time?

Of course, the risk is that Iran realises all this, and then begins to lash out at Saudi and the rest of the Middle East, and try and disrupt all oil production. This is possible, but would probably lead to greater intervention by the US, and would NOT be supported by China - as explained above. Markets have worked all this out I think - which is why oil prices are basically flat on the year.

From my perspective, if US bombing Iran had caused JGB yields to fall, then I might have to have a rethink about markets - but JGB yields still remain in a bear market.

Iran doesn’t matter - and yet the cost of capital continues to rise. Interesting.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.russell-clark.com/subscribe

I have two sons. Of the two, I would say the youngest one is far more money focused. The other day, he asked how I made money. So I told him, I write a substack where people can subscribe for free. I then ask people to pay for the posts that are concerned with investing. On top of that for people who like what they see, they can invest in a fund based on the same ideas. He was suitably impressed by the 15,000 free subscribers. He then asked what the investment strategy was. I felt like the idea of pro-capital versus pro-labour policies would take too long to explain, so opted for the easier option, based on what we have seen so far this year. I said I was waiting for President Trump to crash the markets again. To which he replied, “good strategy”.

What exactly do I mean by getting President Trump to crash the markets? Well the bull market we have seen since 2009, for me, has in many ways had a “private equity” feel to it. What do I mean by that? Well until 2016, it was typical for US fiscal deficits to fall as employment and stock markets rose. “Save in the good times, spend in the bad times” mentality has been replaced with “If we can borrow cheaply, then why don’t we?” mentality. This used to be called running the balance sheet through the profit loss statement. Or short term gain for long term pain. But when does the short term stop, and the long term start?

One feature of this private equity mentality, is that corporate credit is basically seen as good a sovereign credit. Or high yield credit spreads stay at very low levels as the government is happy to ruin its fiscal position to keep corporate financial position in a good way.

The question then is when do people give up on US treasuries as a safe asset? Using the 1970s as a template, it seemed to me when gold starts to outperform treasuries would be the start of problems. It has been brewing for a while, but 2023 really started to move, and with Trump’s re-election gold has soared versus treasuries.

Where it gets interesting for me is when lack of trust in treasuries leaks into problems with the S&P 500. Recently we have seen gold breakout against S&P 500.

From a technical perspective, we had the break higher, and then a pullback in gold v S&P 500. And as I told my 11 year old, I am now waiting for Trump to crash the market, or in other words, this break higher is followed by another move of gold up and S&P 500 down.

How exactly do I think that will happen? Well there are obvious signs of a slowdown everywhere. The Russell 2000 has gone nowhere for 5 years now.

Continuing claims has risen to the highest level since Covid.

What do I expect to happen here? The Fed should come under extreme pressure to cut rates so the US government can keep spending. Check.

And the Trump administration will likely look to stimulate the economy, which would worsen the inflation outlook, so should see foreign investors dump the long end of the treasury curve. That is I would expect to see the curve steepen radically. This is already beginning to happen. Spread between 30 year and 2 year has steepened to 1%. A move back to 3% seems likely to me.

This has always been the problem with “running the balance sheet through the profit loss strategy” - its a one time win. And once its done, it take a lifetime to fix. I don’t think any of the above it is particularly debatable - the question is really just one of timing. The timing looks right to me, in the way that gold and treasuries are moving, particularly in April. I think it was the tremor before the quake. The one reassuring thing about this strategy is that President Trump likes to create a crisis which he can then solve (which gives him leverage). Or what I am saying is that President Trump has an incentive to create a crisis. I just wonder if the next one he creates he is not able to “solve”. Or as my 11 year old said - betting on a crisis here seems like a good strategy.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.russell-clark.com/subscribe

Ever since shale production was commercialised, US natural gas production has surged.

The US produces so much natural gas, that the Permian region can see negative prices for natural gas from time to time. This happens as Permian oil producers also produce natural gas as a by product - so will produce natural gas almost by accident.

Just these two graphs alone make the idea of the US suffering from a natural gas shortage seems crazy. And yet, I see two other graphs that seem to suggest something is up. First of all, Henry Hub natural gas futures are inflecting higher. January 2028 natural gas futures seem to be moving back to a bull trend after a two year consolidation.

And natural gas pure play - EQT Corp, is nearly at new all time highs.

EQT has an arresting presentation - where they are suggesting 25% growth in natural gas demand by 2030.

They also highlight that LNG capacity is going to increase substantially.

Personally, I am not sure about the data centre demand - I thought the DeepSeek news earlier this year would put a dampener on demand - but with Nvidia back at close to all time highs, we can assume those data centres will get built (whether they will make any money - we will leave aside for the moment). In essence, EQT is saying the US will see demand for natural gas rise by 25 bcf/d - with 10bn rise needed in the Appalachian region where EQT operates, and 15bn in the Permian region, where most of the LNG capacity is being built. The problem is that the inventory of wells that can be used to boost production has fallen dramatically.

The growth in US natural gas production has also slowed.

EQT make a couple more interesting points. As natural gas production and demand has increased, there has been no investment into storage. This also has to do with reduced seasonality making storage less profitable.

I think the market is recognising the tightness of the US natural gas market. For old people, I am beginning to get visions of the dot com bust - where surging natural gas prices in California in 2000 led to money being pulled from speculative ventures, and collapsing tech pricing.

Of course this is all just a story - but it is starting to look like a good story to me. And just like in 2000, gold is beginning to inflect against the S&P 500.

The other thing about this story is that I hope it is true. A good bear market would be very good for me - as I love to short sell. This makes me slightly cautious, when people want something to be true, they will convince themselves is true. The one thing that is problematic to this argument would be the rise of renewable energy - but renewable energy really is not suited to data centre usage. I think the best we can say is that it looks likely true, and if it is true, I am positioned to do well.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.russell-clark.com/subscribe

This is a free preview of a paid episode. To hear more, visit www.russell-clark.com

I have recently gone long of two defence stocks. Despite growing backlogs, both still trade below all time highs. Does that mean these stocks are bargains, or does it mean something else? Generally speaking, stocks tend to follow order backlogs - as it gives you confidence that the sales will be there. If we take a topical example - I look at Boeing and think, “wow a company that makes planes that crash - this must be cheap”. I would be wrong. Boeing market capitalisation is still a large multiple of invested capital. If ever a stock was designed to drum up support for anti-capitalism and anti-American fervour this is it. I am disappointed and slightly disturbed when my plane is a Boeing these days.

Why has Boeing stock not crashed and burned, unlike its main product? The truth is that somehow it has still managed to get orders for its product (a duopoly is useful that way). Its order book has returned to new all time highs.

I am interested in defence stocks, which also trade off their order books. Unlike Boeing, they intentional kill people, but at least their is some moral clarity there. One Japanese stock that I looked at this year is Mitsubishi Heavy. It is not a pure play, but does a number of things - from nuclear power plants, to aviation to missile and space systems. It has one of the more extraordinary share price charts - booming in the 1980s, and then sideways for 30 year before suddenly booming again.

Bloomberg gives a 20 year order backlog for Mitsubishi Heavy. It has been a very cyclical business - but judging by the share price, investors are thinking we are moving from a cyclical to secular growth story.

As recently as 2023, the EV of Mitsubishi Heavy was below the value of its invested capital (a very VALUE) metric. Implies that the market think its invested capital will never generate a good return. Now the EV is 3.5 times invested capital.

So how does this compare to the two defence companies we have invested in recently?

I have liked the long gold, short treasury trade for a few years now. But recently, I have seen a lot of chatter about bonds being a buy again, and gold having topped out. It has had a good move - but has been moving sideways of late.

If we are going back to the craziness of the 1970s, then we have plenty of upside left in the tank!

Why anyone thinks 5% on the 30 year treasury is a buy is a mystery to me. During the 1990s and 2000s, governments worked hard to balance the budget. These days they have all but given up. Only much higher interest rates would force any unpopular fiscal policies these days. 10% on the long end still seems likely to me.

What about gold? I think of gold as a hedge on wage and food inflation (I see these two things as interchangable). If I use Japanese food inflation as a guide (its useful because it was flat from 1990 to 2012 - you don’t need to use a log scale), in 1970s, food prices went up 200 to 300%. We have broken higher recently - and I can easily see Japanese wages rising 200 to 300% from here.

Some people say that gold is over owned - but ETF holdings still look low to me.

You could argue that gold has been driven by central bank buying. I would agree with this. But as Section 899, as well as the treatment of Russia foreign reserves should highlight - only an idiot would be buying another countries bonds as a foreign reserve. Gold is now 20% of central bank reserves. For some that feels like enough, but for me, on current political trends - a number closer to 80% seems likely to me.

Much of the current bull market in the US has been built on the Trump and Biden administrations abusing the reserve currency position of the US. The recent break out in gold versus the S&P 500 only just looks like the beginning to me.

So I can understand why people might think GLD/TLT are done - and there are very rational economic reasons to think this. But politically, we are only just getting started I think.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.russell-clark.com/subscribe

This is a free preview of a paid episode. To hear more, visit www.russell-clark.com

Macro analysis, much like movies about the future, has a natural dystopian bias. Why? Because people like to hear about doom and gloom - it sells well. This is a fact. In my investment career - I have made more money being long, but raised far more money when I am net short. The logic of it is easy to understand. If something does go wrong, having a hedge to ease the pain helps - and if you can be net short and make money even better - something I have been able to do for long stretches of my career. What I began to focus on was capital flow and asset markets (and hence the title of the substack). I originally used BIS data - but started to notice that a newly reported economic metric - net international investment position (NIIP) - also seemed to do the trick. I mention this, as almost every bearish article on the US now seems to mention NIIP. This chart from the Economist would be pretty standard.

The thing is that NIIP worked well in predicting the Eurocrisis, showing that German investment was pouring into Spain, so that at some point when the Germans asked for their money back - there would be a problem.

The first problem with a bearish view on S&P 500 is that from 2016 onward, the US NIIP has blown out - but there has been no equivalent increase in NIIP elsewhere. Or in other words, the US seems to be booming WITHOUT any inflows from other countries.

Taking the analysis a bit further, I try to strip out any NIIP that comes from the US dollar being the reserve currency - so separate the NIIP into a private sector and public sector measure. Using the dot com bubble as a comparison, I thought the US looked overvalued in 2016.

What I am trying to say, is that the NIIP backtested well - and I then used it to time being bearish on China successfully, but since 2016, it has totally failed as a macro economic indicator. I suspect that the US NIIP data does not work anymore, because of two interrelated factors - one is massive offshore profit shifting, which NIIP is picking up, and share buybacks, which has shifted returns from income focused to capital gain focused. So if NIIP is a bust, is there any other reason to be bearish on the US? As mentioned, I used to use a different data set to gauge asset flows.

This is a free preview of a paid episode. To hear more, visit www.russell-clark.com

It is hard to believe now, but after the Great Depression, FDR regulated the production of oil so that prices would not fall. The Texas Railroad Commission (TRC) was charged with controlling production to prevent falling prices. As almost all oil was drilled in the Texas, they had great power. The TRC was in some ways too successful, as drilling and exploration slowed in the US, and then in time, so did production, until the shale revolution.

Restriction on oil production were implemented as the US government was seeking to control the deflationary forces of excess competition and excess production that had plagued the country in the 1930s. However the oil shocks of the 1970s ushered in a new “pro-capital” era of management, with substantial deregulation.

I have been wondering if the current rice shock in Japan might usher in a policy change as well. Rice prices in Japan have been stable for decades, but have suddenly surged. The price of “gas” in America is probably as important at the price of “rice” in Japan.

Before I go on to what policy changes this might engender, the surging rice price does make me bullish on Japanese land values. To me Japanese property prices surged well above food prices in the 1980s, and then spent three decades catching down. Assuming wages follow food prices (which they must) then Japanese property prices should do well if rice prices stay strong.

For context, US farm land has been a great inflation hedge - trading closely with gold, even to the extend that farm land weakened in the 1980s at the same time as gold.

Even more extraordinary is that Japanese farm land is now half the price of US farmland.

How has this come to be? Well Japanese rice prices have always been higher than the rest of the world. During the industrialisation boom years, keeping rice prices high was seen as a way of transferring wealth from the cities back to rural areas. The problem was that the government had to come in and buy production at high prices, which they did not want to do, so a policy was instituted to pay farmers to keep land fallow. This has been largely successful, with the area for cultivated paddies and number of farmers declining.

While Japan is largely self sufficient in rice production, Japan has become steadily less sufficient in total food over time. Many vegetables used in Japan come from China these days.

Bit like the oil shock led to deregulation in the US of the oil industry, I am wondering if the rice shock will lead to a deregulation of the agriculture industry in Japan? I am encouraged by the appointment of the Shinjiro Koizumi as Agriculture Minister. His father, Junichiro Koizumi, was a very reform minded Prime Minister, and that most rare thing in Japanese politics - unpredictable and willing to take on entrenched interests.

Most obvious reforms would be allow profitability of farming to rise, and to deregulate the industry. Being reliant on China for food supply seems also a good driver for reform. For me, there has been one stock that I thought would do well on any possible reform - and it is beginning to move.

This is a free preview of a paid episode. To hear more, visit www.russell-clark.com

May was an up month. How much? Well we are just ironing out some issues so waiting for a final valuation. It is taking a bit long as this is first time we are doing the calculation - next month should be quicker. I have written the newsletter, and just now waiting for final valuation. All I know is that we were net short, and made money - so that’s good enough for me.

In a recent update, I said I still saw value in one defence stock in which I took a position in. It has appreciated 20% since I got involved, so being a greedy b*****d, I have started looking for another undervalued defence stock. I think I have found one. There is a lot to like about it - the only question is why has it not moved before? It is possible that no one has noticed as it has zero Wall Street coverage.

European defence is a hot area to invest right now. You can always tell its become a big topic among allocators when Goldman Sachs develops a thematic basket.

In case you were wondering what stocks this included - here is the make up of this index.

The first stock, Rheinmetall, is now EUR82bn market capitalisation (USD93bn market capitalisation). This compares to USD 113bn market capitalisation for Lockheed martin, or EUR 134bn market capitalisation for Airbus, with a duopoly position in commercial aircraft and a very large stake in European missile business MBDA. Or in other words, Rheinmetall is pricing in very serious growth.

The excitement around defence related stocks is pretty easy to understand. The UK, which has always fancied itself as a warrior nation (with good reason), finds itself needing to spend a lot of money. This chart makes revenue growth of 200 or even 300 percent in real terms (let alone nominal terms) seem likely.

The problem for me, is that current expectations on defence spending rests on two assumptions that look problematic. First, Russia needs to stay very hostile. I am not sure this is the as solid as it looks. Russian losses now exceed losses seen by the US in Vietnam, the US in Korea, and more importantly the Soviet Union in Afghanistan. The Soviet Union war in Afghanistan is rarely talked about (other than but aficionados of Rambo 3). The military losses in Afghanistan certainly led to disillusionment with the Soviet Union and sped up its dissolution. One cannot but think that Putin must be coming under similar pressure.

Secondly, Ukraine’s extraordinary raid to destroy large parts of the Russian bombing fleet must count as one of the more historic moments in military history. It will change war as we know it, just at the “Charge of the Light Brigade” signalled the end of mounted military, and Hiroshima and Nagasaki ended the possibility of Great Powers directly attacking each other.

According to the Economist, 150 drones and 300 bombs were used in a raid that destroyed 41 aircraft, which included Russian equivalents of AWACs. Using very rough numbers, Ukraine maybe used USD 1mn of hardware to inflict USD 3bn of damage to Russia. For this reason, I am very cautious on builders of equipment that is non-drone related. The risk is that military budget suddenly and radically get redirected. For that reason, I only like three areas in defence for investing.

Firsts, munition and armaments. We may move from fighters to drones, but they will still need to be armed, and the war in Ukraine has drained inventory across Europe.

Second, countermeasures and jamming technology. The ability to secure areas against drones has become vital.

Third, potentially submarines. And only potentially. The ability to approach a target and then launch drones seems to be vitally important. Could submarines launch drone swarms from offshore, and then be able to leave the area undetected? Given that the drones in the Ukraine war are able to hover for extended periods, and operate remotely using AI, this seems very possible.

Given the ubiquitous nature of drones, it seems all wars will be drone wars going forward. Heavy equipment will still be needed, but the ability to defend against drown swarms will be vital. Jamming and munitions and submarines is where I think the best place to be is. I suspect when drones successfully start attacking aircraft carriers, then politicians will recognise the new world.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.russell-clark.com/subscribe