Insight Focus
Global consumption of dairy as an ingredient has been rising over the past decade. However, totals can be misleading. The world’s population has grown from around 7.5 billion in 2015 to over 8.2 billion in 2025 – an increase of roughly 10%. When ingredient consumption is expressed on a per capita basis, a more complicated and interesting story emerges.
Setting the Scene
Global dairy ingredient markets have expanded in total volume terms over the past decade – more tonnes of skimmed milk, full cream milk powder and whey powder are being consumed as ingredients today than in 2015. But total volumes are shaped by two forces at once: how much each person consumes, and how many people there are. With the world’s population growing from around 7.5 billion in 2015 to over 8.2 billion in 2025 – an increase of roughly 10% – total volumes can rise even as individual consumption is stagnating or falling.
For dairy ingredient producers and traders, the message is clear: total volume growth alone is an insufficient guide to market health. Per capita trends, segmented by category and ingredient, are essential for understanding where real demand growth is occurring – and where it is not.
Fermented Dairy: A Quiet Recovery
Skimmed milk’s role in yoghurt and sour milk products tells perhaps the most encouraging story from a dairy industry perspective. Per capita consumption stood at around 3.20kg in 2015, dipped to a low of around 3.02kg in 2020 – the COVID-19 disruption year – and has since recovered to approximately 3.16kg by 2025.
The COVID dip in fermented dairy consumption was real but temporary. Per capita skimmed milk use in yoghurt and sour milk products has been on a steady recovery path since 2020 and is approaching its pre-pandemic levels.
The recovery is likely driven by a confluence of factors. Firstly, the well-documented shift towards functional and “better-for-you” foods accelerated during and after the pandemic. Secondly, yoghurt benefits from its positioning as a source of protein and probiotics. And lastly, the expansion of premium and regional yoghurt formats – Greek-style, Icelandic skyr, kefir – has introduced dairy into new consumption occasions in markets where traditional fermented milk was not embedded in diet.
There is also a structural change worth flagging. The rapid mainstreaming of GLP-1 drugs is reshaping how consumers approach protein and satiety and high protein fermented dairy formats sit within that shift.
Users of GLP-1 drugs typically eat less overall but prioritise protein density per calorie, a pattern that favours Greek yoghurt, skyr and protein-fortified fermented milks over less nutrient-dense alternatives. As GLP-1 usage grows, we would expect this to support per capita skimmed milk consumption in this category, potentially accelerating the recovery trajectory visible in the chart above.
Drinking Milk Products: The Long Decline of Fluid Skimmed Milk
Fluid skimmed milk’s role in drinking milk products has fallen by more than a quarter on a per capita basis since 2015. This is the steepest decline in the categories examined in this article and reflects a shift in consumer preferences away from low-fat liquid dairy.
The drivers here are reasonably well understood. Consumer attitudes to dietary fat have shifted. The “fat is bad” rhetoric that drove decades of skimmed and semi-skimmed milk consumption has given way to a more nuanced view. This has been influenced by research into the health effects of full-fat dairy, the popularity of higher-protein diets and the mainstream success of formats like Greek yoghurt and full-fat cheese. Full-fat and whole milk have benefited at the expense of skimmed formulations.
There is also a structural element. The markets where skimmed milk in drinking products has historically been dominant (Western Europe and North America) are mature, contracting in per capita terms, and no longer growing fast enough to offset declining per capita consumption within them. Faster-growing markets in Asia, Latin America and Africa have historically favoured different products, including full cream milk powder reconstituted as liquid milk.
Skimmed Milk Powder in Drinking Milk Products
It is worth pausing here to distinguish what the previous chart and the one below are showing. The chart above tracks fluid skimmed milk consumed directly as a drink, measured in kilograms of liquid per capita.
The chart below tracks skimmed milk powder used as an ingredient in drinking milk products (primarily reconstituted liquid milk in markets without developed cold chains) measured in kilograms of powder per capita. The two are therefore on different bases and not directly comparable in their raw form; we return to a like-for-like comparison after.
Skimmed milk powder’s per capita contribution tells a different and more stable story than fluid skim. Consumption was approximately 0.112kg per capita in 2015, dipped slightly in 2016, recovered to 0.112kg in 2019, and has since settled at around 0.110kg by 2025 – a very modest net decline of around 2.3% over the period.
The relative stability here, compared with the sharp decline in liquid skimmed milk, reflects the role of SMP in reconstituted dairy products in emerging markets. Unlike fresh liquid skimmed milk (a predominantly developed-market product), SMP is a traded commodity that is used to standardise and fortify dairy products across a wide range of markets, including those with less developed cold chain infrastructure. Demand has proven more stable as a result.
To compare the two on like-for-like terms, we can convert fluid skimmed milk into its powder-equivalent by applying a skim-solids factor of approximately 9.5%. On that basis, the 2.7kg per capita of fluid skimmed milk consumed in 2025 equates to roughly 0.257kg of SMP-equivalent – down from around 0.342kg in 2015.
Combined with the 0.110kg of SMP consumed directly as an ingredient, total skim solids in drinking milk applications have fallen from approximately 0.454kg per capita in 2015 to around 0.367kg in 2025, a decline of roughly 19%. In other words, even after accounting for the more resilient SMP ingredient channel, the total skim-solids picture in drinking milk remains one of sustained structural decline.
Full Cream Milk Powder in Drinking Milk Products
Full cream milk powder in drinking milk stands out as one of the few ingredients in this article with genuine per capita growth. Consumption rose from approximately 0.115kg in 2015 to a peak of 0.118kg in 2021 and has remained elevated at around 0.118kg through 2025 – an increase of roughly 2.7% over the decade.
The use of full cream milk powder reflects the expansion of reconstituted whole milk consumption in markets across Asia, Africa and the Middle East – where long shelf life, transportability and affordability matter most.
This growth reflects the expansion of drinking milk consumption in markets where FCMP reconstitution is the primary format – across sub-Saharan Africa, parts of Southeast Asia and the Middle East. These are the growth markets for dairy globally, and FCMP sits at their centre. The slight softening since 2021 may reflect some switching towards fresh-equivalent products as cold chains improve in certain markets, but the overall trend is clearly positive and structural.
Chocolate Confectionery: A Category Under Pressure
Chocolate confectionery is the category that generates the most discussion. The dramatic rise in cocoa prices – driven by supply shocks in West Africa and compounded by low stocks – began in 2021 and accelerated sharply through 2023 and 2024. The per capita data for dairy ingredients in chocolate tells a story that is broadly consistent with what those price pressures would predict.
Full Cream Milk Powder in Chocolate Confectionery
Per capita consumption of full cream milk powder in chocolate confectionery was approximately 0.054kg in 2015. It remained broadly stable through to a post-COVID recovery peak of 0.054kg in 2021, then began a sustained decline, reaching approximately 0.052kg by 2025 – a fall of around 4% over the decade and nearly all of that concentrated in the post-2021 period.
The timing is significant. The inflection point in 2021–2022 coincides almost exactly with the beginning of the cocoa price shock. As cocoa prices rose, manufacturers faced two broad options: absorb the cost or reduce the size and/or frequency of chocolate products.
Shrinkflation – reducing pack sizes without reducing the price – has been widely documented in chocolate across the 2022–2024 period. Either mechanism would reduce the total dairy ingredient volume embedded in chocolate consumption per person.
The more recent picture, however, points to a potential inflection. Cocoa futures have fallen sharply from their 2024 peak, retreating towards levels last seen in early 2023 on both the London and New York exchanges. If sustained, this pullback removes the single most important headwind facing chocolate demand over the past three years.
Manufacturers will be slower to reverse shrinkflation than they were to implement it, and price points rarely return symmetrically to pre-shock levels – but the direction of travel is clear.
Our working assumption is that, as chocolate becomes more affordable at the shelf through 2026 and beyond, per capita consumption volumes will recover towards pre-spike levels, pulling dairy ingredient demand up with them. The per capita declines in FCMP, SMP and milk fat shown in the charts above are therefore likely to represent a cyclical trough rather than a new structural baseline.
Skimmed Milk Powder in Chocolate Confectionery
The pattern for skimmed milk powder in chocolate closely mirrors that of FCMP. Per capita consumption sat around 0.046kg in 2015, recovered briefly to a local peak of 0.047kg in 2021, and has since fallen to approximately 0.044kg in 2025 – a decline of around 5.3%, the steepest proportional fall among the chocolate ingredients measured.
SMP is typically used in milk chocolate formulations and in filled chocolates. Its decline tracks the broader contraction in per capita chocolate volume. There is no clear evidence of substitution towards alternative ingredients. Rather, this appears to be straightforward demand destruction driven by higher final prices passed through to consumers.
Milk Fat in Chocolate Confectionery
Milk fat – essentially butter fat – in chocolate confectionery follows a similar trajectory. Per capita consumption was around 0.0188kg in 2015, ticked up slightly to 0.0190kg in 2021, and has since declined to 0.0180kg by 2025, a fall of approximately 4.4%.
Milk fat is used in chocolate primarily to modify texture. It appears in both milk and dark chocolate formulations. Its decline is consistent with the broader pattern of lower chocolate volumes per capita, and there is limited evidence of meaningful substitution by non-dairy fats in mainstream chocolate production over this period.
Conclusion: Two Speeds of Dairy
The per capita lens reveals a market operating at two distinct speeds.
On one side sits fermented dairy – where per capita skimmed milk consumption in yoghurt and sour milk products has largely held its ground, dipped during COVID, and is now recovering. Alongside this, full cream milk powder in drinking milk has actually grown, reflecting the structural demand for reconstituted dairy in high-growth emerging markets. These categories have genuine momentum.
On the other side sits a set of applications – most visibly, chocolate confectionery – where per capita consumption of every major dairy ingredient has been contracting since 2021. The timing and consistency of these declines’ points to the cocoa price shock as the primary driver – as chocolate became more expensive to produce and buy, consumers consumed less of it, and with it, less of the dairy it contains.
Skimmed milk in drinking milk products remains the most structurally challenged: a sustained, decade-long decline of over 26% in per capita terms that reflects a fundamental shift in consumer preferences away from low-fat liquid dairy in the markets that have historically driven its use.
In aggregate, the data suggests that dairy ingredient consumption is holding its own globally on a per capita basis, but perhaps only because growth in emerging-market applications is offsetting structural declines in developed-market categories. The headline tonnage figures, driven by population growth, overstate the underlying demand momentum.
Looking forward, the interplay between these trends matters more than any single category in isolation. If our working assumption on chocolate is correct — that falling cocoa prices allow per capita volumes to recover towards pre-spike levels – then dairy demand from confectionery should begin pulling in the opposite direction to its recent trajectory, supporting FCMP, SMP and milk fat simultaneously.
Layered onto the existing picture, this produces a nuanced aggregate: skim solids continue to decline in drinking milk faster than they recover in fermented dairy (implying a net per capita fall in total skim demand), while FCMP benefits from the double tailwind of continued strength in emerging-market drinking milk applications and a chocolate demand recovery.
The geography reinforces this. Population growth over the coming decade is concentrated in sub-Saharan Africa, parts of South and Southeast Asia, and the Middle East – precisely the markets where FCMP-reconstituted liquid milk is the preferred format and where chocolate consumption, while still small on a per capita basis, has meaningful room to grow.
The product best aligned with where people are being added to the global population is also the product best positioned for cyclical recovery in confectionery. Of the major dairy ingredients examined here, FCMP has the most compelling demand story on both a structural and cyclical view.
This leaves an open question on the skim side. If fluid skimmed milk consumption continues its structural decline, the skim solids displaced from that channel must go somewhere. The most obvious destination is the dryer – which would imply rising SMP production (and, by extension, related products such as MPC) not because demand is pulling it through, but because supply is pushing it out.
On current evidence, skim demand across drinking milk and chocolate is weak and only partially offset by recovery in fermented dairy. Combined with the prospect of additional solids seeking an outlet, this points to continued pressure on SMP prices and raises a genuine question about whether regions not traditionally focused on full cream powder – notably the US and EU – may find themselves producing more FCMP simply to rebalance their product mix. How that plays out will be one of the more interesting structural stories in dairy over the next five years.
