One month ago, in his latest letter to clients, Horseman Capital's Russell Clark revealed a new "investing" strategy using ETF flows as a catalyst for positioning and bets.

Citing the transition from active to passive as a catalyst that makes markets increasingly more inefficient, something One River's Eric Peters noted in a recent weekly note, Clark repeated a lament made by many short sellers, stating that there "are complaints from some quarters about it being harder to short sell as flows of money push up stocks."

So what is his new shorting philosophy? This is how he explained it, using his biggest short at the moment, retail REITs:

The biggest short sector in the fund are REITs. In the US, they are mainly retail REITs, and there are two reasons for this. One is that we have guaranteed sellers in the Japanese US Reit fund. The other reason is the appalling performance of the major tenants. However, as an aside, I like them as a short area as they have the highest exposure to ETFs of any sector.

Bloomberg allows you to find the biggest ETFs and open ended funds which are invested in US Real Estate Sector. The top 28 funds have total assets of 187bn USD, of which 13.3bn USD invested in Simon Property Group, that is 24% of Simon’s market cap. However, Real Estate passive funds are not the only passive fund invested in Simon. When all passive funds weights are added together I get over 50% of Simon Property Group shareholders are passive. I wonder who will become the buyer if all these funds start to see redemptions if there are some problems in US commercial real estate?

The long bull market in passive investment has made them willfully blind to the liquidity risk that they are running. Passive investments are concentrated in the US market...

And, if Eric Peters is right, "when these markets do finally have a correction there will be no bid for many of these stocks", so all Clark has done is tighten the universe of ETF unwinds from the entire market to a market sector or subset of stocks, in this case the retail REIT space.

What was most interesting about the new Horseman approach, however, was that it combines fundamentals - in this case the declining purchasing power of the US consumer and the secular shift to online buying - with market inefficiency in the form of ETF flows that have pushed stock prices ever higher from their "fair value" in anticipation of an eventual sharp move lower as ETF inflows finally reverse. That said, it was not immediately clear what the catalyst for this reversal in ETF flows would be.

In any case, we said one month ago that one can repeat the exercise for all other sectors, and stocks, that have a substantial exposure to ETFs, and slowly but surely the shorts will start to accumulate, putting further pressure on sectors and stocks that have been abnormally influenced by passive flows, until finally the money flow support breaks, leading to a crack in the current market topology, potentially followed by the next market correction, or worse.

Now, courtesy of Goldman, we have the full breakdown of the most and least concentrated sector ETFs. As the chart below show, the five most concentrated ETFs currently, on both a relative in terms of current weighing of the Top 3 stocks, and absolute (in therms of overall weight of the top 3 names) basis, are the Consumer Discretionary (XLY), Info Tech (XLK), Financials (XLF), Energy (XLE) and Utilities (XLU), all of which have never seen a greater relative weighing of their top 3 companies. On the other end are the Healthcare (XLV) and Industrials (XLI) ETFs.

For those who think the logic behind the Horseman ETF (out)flow-based trading strategy works, the best trade would be to go short all the most heavily weighted ETF constituent stocks, while shorting the least concentrated ones, in creating a relatively low-risk pair-trade ahead of the next "August 2015" ETF Flash Crash, which is absolutely assured to take place again, the only question is when, and who to keep the trade on with the lowest possible negative carry.