Join us for our 4th global annual Innovation Series Register now
Logo
Exploring correlation in CPP freight and arb indicators

Exploring correlation in CPP freight and arb indicators

This summer has seen diverging trends in Atlantic Basin clean freight rates, as a reaction to key demand centers importing more or less than expected.


Mick Strautmann
Mick StrautmannMarket Analyst

This summer has seen diverging trends in Atlantic Basin clean freight rates, as a reaction to key demand centres for clean products importing more (diesel) or less (gasoline) than expected. There are numerous factors behind sharp freight rate movements, including product demand, vessel availability, transit disruptions and market sentiment, to name a few.

By looking at vessel availability and product exports in isolation across three major clean product routes, we find that vessel availability exhibits stronger correlation with rates than exports, and that differences in the availability to rate relationship across routes can be due to different availability cycles.

For the following clean product routes, we gathered weekly data on exports (Aug-22 to Jul-25), 7-day ballast availability in the respective origin shipping region, and freight rate indices created with Vortexa’s Anywhere Freight Pricing (liquidity weighted rate across the top-five port to port routes for the dominant vessel class on each of the three routes).

Under the assumption that despite numerous omitted variables (product demand, market sentiment, disruptions…) there is a linear relationship between exports, vessel availability and rates, we computed Pearson correlation coefficients across lags of six weeks before (t=-6) to six weeks after (t=+6). The series are first stripped of seasonality by using a four week moving average filter, so that the correlations focus on short term fluctuations rather than seasonal trend. The three cases below assess how vessel availability and export numbers at different weekly lags correlate with rates in the current week (at t=0).

Diesel from PADD 3 to Northwest Europe

Diesel PADD3 NWE

Pearson correlation by lag (weeks) between PADD 3 diesel/gasoil exports to NWE, Gulf of Mexico MR2 availability and respective MR2 freight rates

The PADD 3 diesel/gasoil to Northwest Europe route has been very active this summer, amid high European diesel demand and favourable arbitrage economics. The peak correlation on this route is exhibited between vessel availability and rates (r = 0.35 at week 3), consistent with an MR2 availability peak about three weeks after a strong rate move. As vessels are usually fixed several weeks in advance, this is in line with a reasonable repositioning timeline. More generally, the results suggest that higher rates can trigger higher vessel supply in weeks 1-5, while lower availability in weeks -2 to -1 is supportive of rates. 

Exports on the other hand only exhibit a weak peak correlation at t=0 (r= 0.22), suggesting to a limited extent that rates move in accordance with intra-week changes in exports numbers. Correlations are mostly positive from week -3 through week +3, suggesting that higher exports both in the recent past and the near future tend to support current freight rates.  

Gasoline from Northwest Europe to PADD 1

gasoline nwe padd1

Pearson correlation by lag (weeks) between NWE gasoline/blend comps exports to PADD 1, NWE MR2 availability and respective MR2 freight rates

US Atlantic Coast gasoline demand has been underwhelming this summer driving season, with PADD 1 gasoline arrivals from Northwest Europe in July at their lowest level in three years. The chart above shows us that MR2 availability in Northwest Europe exhibits a much stronger correlation with rates than export levels. Peak negative correlation occurs at week 0, highlighting an almost immediate negative relationship between availability and rates. This summer, given high MR2 activity on the PADD 3 to NWE route, the NWE vessel balance has been oversupplied, and the peak above could indicate that every additional MR2 arrival in NWE brings downward pressure to the NWE-PADD 1 MR2 freight rate within a week.

Middle distillates from Middle East Gulf to Northwest Europe

mid disti MEG NWE

Pearson correlation by lag (weeks) between MEG middle distillate exports to NWE, MEG LR2 availability and respective LR2 freight rates

The Middle East Gulf is the main supplier of clean products to Europe, but transit times, especially since Red Sea disruptions began in late 2023, are generally longer than from PADD 3. We can see in the chart above that availability correlation to rates becomes more pronounced from two weeks ago until now, reaching a peak negative correlation of -0.36. This line reaches a positive peak at week 4, highlighting that higher rates today could signify a peak in vessel supply in four weeks. 

Although less significant, a similar pattern can be observed on the export to rate correlation series, taking into account that these series are likely themselves correlated. This shape at weeks -2 to 0 could then indicate that if vessel availability drops faster than the relative export decrease, rates today will tend to be higher. Meanwhile, high freight rates now are only a weak but consistent indicator of higher flows to come in weeks 1-5.

Cross-route comparison shows availability as a consistent pattern

all routes corr

Pearson correlation by lag (weeks) between availability and freight rates (LHS) and exports and freight rates (RHS) across three routes

Across all three routes, vessel availability consistently aligns more with freight rate swings rather than exports. Differences in the correlation strength across lags can give an insight into the cyclicality of vessel availability, and how freight rate moves can affect these cycles. For example, the left chart above shows that high freight rates today across all routes can lead to different vessel repositioning timelines. 

The consistent pattern appears to be that availability in the past two weeks up to the present is negatively correlated with freight rates, while future availability is positively correlated, as vessels reposition to benefit from more attractive markets. 

There is an indication of an eight week cycle in the availability-rate relationship, with four weeks each from peak to through and back to peak. With some imagination, such cycles can also be seen on the export/rates chart, which are however much shorter with about four weeks from peak to peak. While the specifics are challenging to untangle, it is clear that freight and arbitrage markets are subject to a high level of cyclicity, with price signals not only closing opportunities but often even triggering over-compensation, thereby reversing the market situation from undersupply to oversupply and vice versa, for both vessels and cargoes.

JOIN OUR NEWSLETTER

Loading form...

About the analyst

Mick Strautmann

Mick Strautmann

Market Analyst