Capital Flows and Asset Markets

Generally speaking the US government fiscal position has looked terrible. On a 12 month rolling basis, for the last 15 years despite full employment and record asset prices, revenue to the US government has never looked close to meeting outlays.

Which has fuelled an ever large amount of federal debt to GDP. Inflation, and soaring stock markets have usually led to falling debt levels, but in this case, it continues to climb.

President Trump has eschewed obvious options like taxing private equity, or progressive taxation for DOGE and tariffs. Neither seemed remotely likely to balancing the budget. But when we look at the monthly summary, June saw a 27bn surplus, compared to 71bn deficit in June of last year - a close to USD 100bn improvement.

Outlays in June fell by USD 40bn compared to last year. I am struggling to find where that fall comes from. All the major items show continued growth in spending. Revenue was up an impressive USD 60bn compared to June of last year. The vast majority was from higher individual income tax. Corporate tax take was down, but custom duties was up USD 20bn from a year ago.

How high could this number go? Well the US is importing around USD250bn a month steady state, so we can simplify and say there was an effective 10% tariff in June.

President Trump throws around a lot of number on tariffs. I am going to “guestimate” that maybe he can get to 30% tariffs before you get real substitution (buying US made rather than imports). This would give a USD75bn monthly bump to revenue. Lets also assume it causes a 15% bump in wages, so income taxes also rise by 15%. That would add another 30bn to tax revenue. USD 100bn increase in monthly revenue to the government is nothing to be sniffed at. Cutting expenses is hard, as the short lived DOGE seemed to prove. One option Trump wants to explore is cutting the net interest expense. This has ballooned from around USD 15bn a month to USD 80bn a month.

If the Federal Reserve could cut rates back to 1 or 1.5% and get monthly net interest spend back to USD 20bn a month, that would also help. What happens with 30% tariffs and 1% Fed Fund Rate? An increase of USD 100bn in revenue (custom duties and higher wages driving higher income take), and 1% Fed Fund Rate would give 60bn cut in expenditure. 160bn a month, over 12 months would be close to USD 2 trillion, which is pretty much exactly what the Federal Government Budget gap.

I assume 15% wages increase and 30% tariffs and 1% interest rates would be pretty inflationary, so maybe US GDP grows by another 25% in the next 4 years. (It has grown 50% in the last 5). This would give USD 37.5trillion GDP, up from USD 30 trillion today.

Assuming the budget gap is fixed then the debt pile should not need to grow anymore, something that last happened in 1999/2000. If that was the case, then government debt to GDP would fall to 85% of GDP, and potentially lower if inflation is higher.

We have already seen something similar in Greece, where decent tax collection combined with inflation has reduced debt levels substantially.

All of this seemed fanciful a few months ago when Trump first announced his tariffs - slumping markets pointed to recession and falling tax collection - but with markets bouncing and asset prices surging, maybe it is doable.

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

When Deepseek turned up earlier this year I thought we had seen a top in Nvidia share price. After a period of weakness we are back at highs, and the world's first USD 4 trillion market cap company. As I understand it, Nvidia has a 90% share of GPUs, TSMC has 90% share of cutting edge chips and ASML 100% share EUV cutting edge lithograph machines. Oddly enough, ASML just warned and has flatlined its share price since 2021, while Nvidia is up 8 times. I find this odd - is there unfettered demand for cutting edge chips or not?

Why the disconnect?

The release of ChatGPT has been transformative to say the least. How does this effect me I wonder? I have an AI agent on my landing page at www.brumbycapital.com which I have been experimenting with. Much like almost all computer programs - it is great at repetitive chores (I have loaded it up with my last 100 substack posts - so it can tell you my recent thoughts on anything). But plainly it does not have access to all the different experiences that have come to effect my thinking. So I could make an AI agent that communicates like me… but not think like me.

But as I learnt a long time ago. Original thinking is all good and well, but if no one knows you are original then wants the point? What I am trying say, is that an internet that is moving from Google centric to AI centric may mean I need to change how I operate. Does ChatGpt mean I need to change my marketing? For the first time since Google came to dominate search there is now a real competition in the search space. When you Google search “Russell Clark” I am the first hit after a biography of a dead actor. When I first started out - I was a second page result at best!

But on Chat Gpt… nothing.

Deepseek was good enough to have me on the front page… Sadly it called me a Kiwi manager based out of Tokyo - not true. Also a big short seller of Tesla… also not true

Of course if you really want to find me you can. But the biggest effect of ChatGPT is that click through rates to underlying websites will get affected. Google offers up links. AI basically tells you all you need to know, and does not prompt you to other websites. This change has already effected clickthrough rates to underlying websites. The Economist shows this.

This makes a lot of sense to me. Clicking through to websites from a Google search is an exercise in dodging pop ups and banner ads, while Chat GPT gives you an ad free summary. Asking it how I invest gives a decent summary.

No live links to Brumby Capital or www.russell-clark.com there. This does have real world consequences. Reddit has lagged the market this year. There have been other losers as well.

There have been other losers from this as websites find less and less traffic. Is this a problem for me? Well according to Substack I am still seeing rising traffic

Paying subscribers has dipped recently - but I tend to see an uptick in bear markets and a drop down in bullish markets - which makes sense to me.

The risk is that text based searches for me get throttled at the source with AI. Intriguingly AI struggle to scrape videos. Transcripts of the podcast I do with posts also tends to be full of errors - so I can see why AI does better with written matetial rather than video or audio. This does make me think that I need to move to more video presentations - and to start looking at posting more to Tiktok. Why? My original business model was always to produce a substantial amount of free content to draw people into either the paid content or investing in the fund. But if AI is going to slow that flow - or break the links, then I need to alter how I market. One option would be to use TikTok more. I have posted a a few videos on Tiktok already - and when I search for myself in Tiktok it throws up the right videos. And, as everyone knows, you can add links to the bio as well.

Why TikTok or not Youtube or Instagram? What I like about Tiktok is that its ads tend to be skippable. I also trust the TikTok algorithm more than I trust Youtube or Instagram. I feel like my videos are more likely to be presented to people interested in my topics. Who knows but an AI centric web means I need to change my style slightly. And TikTok looks like the answer.

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

A well spent youth travelling around the world in my early 20s meant that I was very old to join UBS as a graduate trainee at the age of 25. I was even older to start as a fund management research analyst at the age of 27. I was extremely hungry for success, and I would spend all my time studying what the successful fund managers at my firm were doing. At the time the most successful manager at my firm as a chap called Andrew Green. He came to work maybe two or three days a week, and when asked about his investment ideas or philosophy, his answers were positively cryptic. He was the first, and probably truly most successful “contrarian” investor I have ever seen. Below chart is taken from an article announcing his retirement.

What you should notice is that he really began to outperform during the dot com bust, after underperforming during the dot com boom. Assets under management collapsed during the dot com boom, but as soon as the bubble burst, his style took off. He never really spoke to me (why would you speak to young analyst working for a different team), but I got on well with his analyst, and I asked him how did Andrew found his investment ideas? He told me looked for assets that were basing out of long bear market in either relative or nominal basis, and would take a small position, and then ramp it up as they began to outperform. It was this line of thinking that led him to be long gold miner DeepRoot Durban in 2000.

It was also this type of thinking that led him to be long Thailand, and particularly Thai banks in 2000. These were very deep contrarian ideas at the time.

For me, I started to think about capital flows and asset markets (the title of this substack). Did the capital flows out of the US due to the bursting of the dot com bubble drive gold and Thai outperformance? Or was the improving outlook for Asia and commodities cause capital to be attracted to these assets? Or was the Truth somewhere in between? I never really got a definitive answer, but I do pay a lot respect to inflections are extreme points. It is why I put so much emphasis in the potential inflection of gold versus the S&P 500, which would be similar to an inflection 1929, 1970 and 2000.

Getting away from gold, it is hard not see inflection points everywhere. Josh Hartnett from BoA has shown that we have suddenly reached a new high in European equities versus bonds.

To go along with a similar move in Japan. From 1989 to 2013, you made more money in JGBs than you did in Japanese equities. 24 years of basically the complete opposite of modern money management theory.

This may seem to imply that it is time to get long bonds and short equity - as Andrew Green used to do, but that type of thinking would have made you bearish in 2016, 2020, and this year on US equities.

But it is hard to not get the feeling that equities have become extremely expensive versus bonds. The problem here is that I still rather like the look of gold versus US bonds. Or in other words, I don’t really like bonds here either.

And this is inline with a new cycle high in 10 year JGBS.

The most obvious story here is that makes sense to me is that government finances are in a mess - but the public will not accept any austerity. So at some point taxes on corporates and wealth or tariffs become the only option. And this creates a steady flow out of equites into gold, as we shift from wealth creation to wealth preservation. This seems inline with the Swiss Franc closing in on all time highs versus the US dollar. This graph is log scale.

That makes sense to me - but the question is what level of bond yield causes this political change? At what point do governments bow to the inevitable and move back from regressive taxation to progressive taxation? I find it odd to be a contrarian when record asset prices and record government deficits seem to scream out for some sort of progressive tax regime. I think this is what gold and Swiss Franc are telling you. Time will tell.

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

The vast majority of fund management theory rarely survives contact with market reality. Back testing various theories would have told you to sell the American market years ago, and here we are today, pushing new highs. Fund managers, facing performance and career pressure will almost always become index huggers of some sort, and then work out various justifications after the fact. For this reason, passive investment works, and tends to beat active managers. In a trending market, just blindly sitting there holding the index makes total sense.

In my case, I develop theories, and apply them until the market tell me otherwise. With Brumby Capital, I am currently applying two theories that I have developed myself over my career. The first is a practical one, that allows me to run net short until a bear market happens (if they still happen), and the second one is a macro theory, that should generate a bear market. As it happens, the first theory which is centred around the idea of a “correlation break” is working well, so I am pleased with that. The second one, build around the idea of rising cost of capital, is sending conflicting messages - but I think that is creating an opportunity, rather than telling me wrong. If these conflicting messages resolve itself in a lower cost of capital, then I am wrong, and I move on (to the beach and golf course). If these conflicting messages resolve itself in a higher cost of capital, then I am right, and I make good money. Currently, I am sitting in no man’s land.

When I first heard about Modern Monetary Theory (MMT), to me it just sounded like the sort of policies we see in emerging markets all the time. Turkey has been running a version of it for some time. MMT seems to endorse ever increasing government spending - something that Turkey has been doing for awhile now. Below is a log scale Turkish government spending.

In essence MMT endorses the idea of keep spending until something bad happens. In some ways it has been surprisingly effective, with Turkish GDP rising strongly in recent years, after a decade of stagnation. Note this in USD terms - so “real” growth, not just currency debasement led growth.

The classic monetary effect of this has been for the Turkish lira to collapse in value, again this graph is in log scale. When I first started in emerging markets in 2002, it was around 1 million Turkish Lira to the USD. There was a reform, where 6 zeros were taken off the currency to get it back to 1 Lira per USD in 2003. Lira generally loses value but from 2021, the process has accelerated, going from 8 to 40 today.

The other effect has been for 1 year interest rates to rise to over 40% in Turkish Lira terms.

The net effect on Turkish equities of all this government spending, weak currency and high interest rates is that they have done nothing in US terms for over 20 years.

There is no surprise where MMT policies in Turkey come from. President Erdogan dominates Turkish politics and has pushed the central bank to lower interest rates constantly. For me, a weak currency, high interest rates, and poor equity returns are exactly what I would expect from modern monetary theory. And yet the US has not really followed suit. US Federal Government spending has had a step change higher as it did in the 1970s.

And much like Turkey, nominal GDP has soared - up nearly 50% from 2020.

Has the US dollar really weakened? In gold terms you could argue the last two years has seen the dollar weaken significantly.

But Fed Fund Rate has stayed at historically low levels.

And in gold terms, US equities have done fine.

So why does MMT work in America and not in Turkey? Is it purely political? Erdogan came to power in early 2000s, and took some time before he became very heterodox in economic policy and equity returns became poor. Looking at Turkish CDS - it was in 2018 when markets really started to price risk into Turkish assets. That was around 15 to 16 years after Erdogan came to power.

Trump has been the dominant force in the Republican party since 2016, so coming on 10 years, with at least another three and half years to go. Maybe that is the question. When does the market believe that MMT style policies get embedded into the US body politic? At what point does faith disappear? The best long term measure I can find is S&P 500 versus gold. It seems to inflect at critical points in American economic history - and could be inflecting now.

I think a lot of the market push for MMT came from the example of Japan, who has run large fiscal deficits for years with no effect on the bond market. This was true for decades, but is no longer true now. 30 year JGB yields have just hit a 30 year high.

To me, MMT probably does imply a non-stop bull market in gold. But I would expect equities to suffer. Perhaps it just takes time - but with JGB yields at 30 year highs, and gold at all time highs - perhaps that time is now.

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

Long time readers will know that I “despise” the clearinghouse model. Despise is a strong word - but justified in this case. The move to central clearing was means to make the system more robust, and remove “too big to fail” moral hazard from the financial system. The opposite has occurred, which the system more prone to failure, and requiring almost constant bail outs, with profits largely accruing to hedge funds. I have a long series of diatribes notes on clearinghouses here.

Due to regulatory change, central clearing has spread its ugly little model to almost everywhere in finance. I first worked out how dumb this model when reading this note from BIS on how a Nordic Clearinghouse run by Nasdaq went bust. Essentially clearinghouses have no pricing ability, and no need to risk capital, so just price risk as if today will be like yesterday. They are also incentivised to maximise profits, so then have tendency to be “optimistic” on pricing. What happens is they basically try and offer as much leverage as possible to buyers of risk. The problem is that when there is a shock, everyone then become a forced seller. We recently saw that with US treasury market in April, when treasuries fell even as the stock market was weak.

Why does this happen? The profit maximising efforts of clearinghouse means that eventually markets will become bifurcated between natural buyers and natural sellers.

So who are natural buyers of treasuries? Well until recently Central Banks would buy treasuries as foreign exchange reserves - but they have wised up to the foolishness of that in a Trump America.

These days hedge funds are natural buyers of physical treasuries. Off the run (treasuries that have been issued for a few months) have a higher yield than futures. So a classic basis trade is to buy physical bonds, and then repo them aggressively to make a “risk free” return. Data from DTCC shows this has been surging.

Now what happens when banks have repoed as much as they can to satisfy hedge fund demands? They will start to rely on Central Banks. For the end of June (which is the first quarterly end date since the April sell off in treasuries) we see a big increase in repo agreements with the Fed.

To put this in context, the Fed made changes to dissuade banks from using this facility during Covid - but it is interesting it has made a return recently.

Likewise, the Fed is adding cash to the system again. In essence public money injection to keep hedge funds and banks profitable and credit worthy.

Why do I care about this? Well for me, the obvious mispricing in market remains US treasuries - which are the cornerstone of modern finance. I have been quite clear that I think long dated US treasuries will end up with a 10% yield.

The bear market case for markets is a simple one. When does the Federal Reserve react to buy back credibility? 5 year treasuries have been steady at 4% now for a few years. I also think this goes to 10% at some point.

So clearinghouses, by creating hedge fund demand for treasuries are the weak link in the markets. Clearinghouses are the equivalent of AIG in the lead up to the GFC - the braindead lenders keeping the whole party going. The problem with the bearish clearinghouse view is that they essentially tie the entire financial system together into a single balance sheet - which essentially means when clearinghouses get in trouble, all banks get in trouble, and the US government will bail everyone out. And this is the reason I also like gold.

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 very MIXED feelings about June. So the core trade of GLD/TLT finished the month on lows. So even though this trade still looks good to me, it was a headwind in June.

I also think long gold/short SPX makes sense here, but again this finished the month near lows, so was another headwind.

I also think the cost of capital should be rising for US corporate in particular. One proxy for this is HYG. During the month it traded to a new high - so was a real headwind for the short book.

Leveraged loan index also traded up in June.

If you told me this was how the month of June was going to be for GLD/TLT, HYG and leveraged loans at the beginning of the month, I would guesstimate a loss of 2 to 3% for the fund. That we ended flat is good sign. But with core macro trading this way, I am always a bit worried. And that is why I have mixed feelings.

So what did work? In a month that had a decidedly “risk on” feel to it - the Swiss Franc hit new cycle highs. As a broader dollar weakening trade this makes sense to a degree - but I feel a rising Swiss Franc is a bearish signal.

On the stock side, we had some good success with some longs and surprisingly with some shorts!

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

In my previous post, I lamented how momentum investing has become the most successful style of investing in the last 10 years, after having a decidedly mixed outcome from 2000 to at least 2012.

Michael Green (who writes the “Yes, I give a FIG” substack) emailed me to say basically its a bubble driven by passive flows. He attached a link to this report, “Ponzi Funds”. I have know Michael for over ten years, and been reading his thoughts for longer. I won’t lie - some of Michael’s ideas and language makes my brain hurt, which I guess is a good thing - but I always worry I may have misunderstood something. With that disclosure, when Michael and the Ponzi Fund authors finger passive investing for the continued success of momentum investing, what they are saying is that passive flows push up prices, which attracts more money, which pushes up prices, and attracts more money and so on so forth. In essence we are in the middle of huge, decades long pump and dump. At the corner stone of this view has been the rise of passive investing at the expense of active investing.

The problem I have with this analysis is that correlation does not imply causation. Yes, the rise of momentum investing has coincided with the outperformance of momentum investing, but from my perspective you could argue that investors have seen how poorly active fund management has performed relative to passive, and decided to follow performance. I know for my self that outperforming in a bull market it tough - but outperforming in a bear market is relatively easy for me. The problem is that I was assuming a bear market would come every 7 to 10 years - in line with recessions in the US. Ignoring Covid, which I think we should, we are now in year 17 of the expansion (d’oh). That being said, I think we can tease a bit more out of Michael’s analysis. So passive flows can drive prices? I think this is true. A good example was the rise of ARKK a few years ago. In 2020, there definitely seemed to be a correlation between flow and performance, and when flow stopped and slowed, ARRK underperformed. This is a definitive example of what Michael is talking about.

The problem here is that ARKK was more an actively managed ETF, than a true passive fund. If I look at the iShares Momentum ETF, I don’t see any correlation here either.

So part of the problem I had with the “passive is everything” argument is that the market is usually pretty good at taking advantage of crazy inflows. Remember Gamestop? That did not work out well in the end.

But I do think passive investing is working for another reason. That Michael is right for slightly different reasons.

It is funny to hear my non-finance friends start worrying about their jobs. AI has been infecting finance for decades. Algorithmic trading, AI programs that analyse releases in an instant, and even Bloomberg Industries, that collates industry data has replace my earliest job in finance - collecting data. One thing that became apparent to me is that the rise of AI made investors very reluctant to give money to strategies that were too simple. The simplest, and always a favourite with “traders” everywhere is momentum investing. Below is the Morgan Stanley US Momentum Long Index. I started in markets in 2000, and momentum investors were taken to the wood chipper then, and then again in 2008. Around 2015, I thought momentum investing was done, and then again in 2021 - and I was wrong. I did think in April we were done with momentum investing again, but here we are, back at highs.

Prior to 1999 when this index started is that we would have seen at momentum investing in the Nikkei bull market over the 1970s and 1980s, and potentially Asian markets in the 1990s, and of course in 1920s. Looking at all this periods, I found that when capital flows were extreme, I assumed they would reverse at some point. NIIP was a useful way of seeing this - for example German versus Spanish NIIP became extreme just before the Eurocrisis.

Using NIIP even worked well for Korea in 1998 and again in 2007. I strip out public sector foreign exchange positions to get a cleaner position.

Applying this to the US, at the top of the dot com bubble, private sector NIIP also became extreme, at negative 10% of US GDP. This would have given you a bearish signal in 2016, and you would have been wrong.

So cautious habits I learned from looking at macro when I was younger have stopped working . But as someone who tries to be through, I always looked for confirmation from the bottom up. I try to look at companies in similar industries to identify changing trends. One of my best ever short ideas was iron ore. From 2000 to 2006, Chinese steel companies and iron ore companies moved together. From from 2010, Chinese steel companies were very weak. For me, the weakness of Chinese steel companies meant that the upside to iron ore was capped. In 2008 and 2011, Vale was a momentum trader favourite. Catching this inflection in momentum investing probably made my name.

Using these same methods, I was looking at the tech space, Samsung, which is both a rival and competitor to Apple, had begun to roll over in 2021, which made me think there was downsize risk to Apple. This has not really been the case - the gap between Samsung and Apple has just gotten wider.

You seem similar disparity opening up between Nvidia and AMD for example.

Watching Disney, would have been good for shorting Netflix in 2021, but has not worked since.

All of the above is in some ways versions of Dow Theory - which is saying the best bull markets raise all boats. Or if you look at the Russell 2000, it has been going sideways for 5 years. Traditionally a bearish signal. I have also been thinking the break out gold versus the S&P 500 is bearish.

What would be ideal is if momentum investing started to fail. As in 2012, that is the chance to really make my name again. So what are momentum investors in? Fortunately there is a USD 17.5bn Momentum ETF - MTUM US - that gives us an idea. It is at all time highs having rallied 31% from April lows.

Intriguingly, when I look at it versus gold, it is right back at the 200mda - so decision time for markets. It is weird, having seen momentum investing work for so long, I fell like it momentum investing must continue on, but a dispassionate look at this graph suggests a trend change is afoot.

If you are wondering what is in it, it is pretty transparent with its holdings. The median market cap of the below is nearly USD 300bn, with a dividend yield of 1% and price to book of 6.

Part of my argument is that the move in gold is a sign that the market is beginning to notice that the US government is essentially bankrupt, and that at some point this leaches in to economic problems. I also thought the rise of cheap competitors to US in AI and EVs would put downward pressure on stocks. If I am right, then you are getting another chance to sell stocks and buy gold. If not, then who knows where this ends. Either my old habits are a trap - or I have somehow managed to survive to an ultra extreme point in markets. The markets will decide.

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