The Flostock Laws of Demand
Plan is to write a book about Demand, with chapters that follow the Flostock Laws of Demand. The insights behind it have been developed during 25 years in the industry, combining some practical forecasting experience, some commercial feeling sharpened at the flea markets of Asia, a bit of playing with system dynamics and a slice of common knowledge. Most of it is as simple as Pythagoras, and the power is in the total. These laws have been published periodically in the Flostock News. Here follows a selection:
1 Supply chain is a tube
Basic in Flostock’s view is that there is a direction in the real economy: All goods flow from their basic sources, such as mines, agriculture or oil-fields, to the end consumer. In this flow no product gets lost: all matter that goes into the supply chain also leaves the supply chain. This has important consequences. This we call --a bit presumptuous perhaps -- Flostock’s First Law. It says that if no product exits at the end, no product can be put in at the beginning. The supply chain is a tube, and if it is clogged at the end, it stops. If it is clogged at the beginning, so for a while no new product can enter at the beginning, the chain will deplete. If it is clogged halfway, downstream will deplete and upstream will overstock, until after a while both ends will stop. As a consequence of the First Law, without clogging, if the end market grows 1%, the upstream supply also must grow 1%. This makes it possible to derive a growth trend in upstream demand from downstream demand.
2 Amplitude of the Inventory Wave is proportional to Stock Depth
Companies maintain stock proportional to their turnover, because they need inventory to maintain the sales flow. When demand goes up, inventories go up, and when demand goes down, inventories go down. This creates an Inventory Wave which is super-posed on top of the basic demand curve. Flostock’s Second Law states that the total inventory in the chain, the so-called Stock Depth, is proportional to the amplitude of the Inventory Wave. The inventory amount of one company is generally described as inventory coverage: enough stock for a number of days’ sales. Knowing how many companies are in the supply chain between you and the end market, and estimating the stock coverage in each intermediate firm, you can calculate the total inventory coverage of the chain. From the perspective of the basic industries this so-called Stock Depth can be as long as a whole year. In an example: when end market consumption grows 4%, the whole chain wants to grow their stocks 4 % to maintain the desired coverage. If your Stock Depth is 9 months, this will result in another 3 % of a year in extra sales to build up the inventory in the chain.
3 Amplitude of Inventory Wave is proportional to Change in Growth
If an end market is more volatile, upstream demand will see bigger swings. The reason is that the chain adapts its inventory to demand, so if demand changes quickly, the inventory needs to change quickly. The cumulative effect of demand and inventory adaptations is what an upstream supplier sees. Flostock’s models do take this volatility and the response of the chain into account, and therefore become more reliable in volatile times, contrary to most competing systems. Sales manager-driven forecasting, no matter at what level of sophistication and supply chain collaboration, cannot do this. Statistical forecasting goes wrong at every peak. The huge forecasting systems used by banks and governments all but crash when markets make a turn. In a world that is increasingly VUCA (Volatile, Uncertain, Confusing and Ambiguous), stock & flow based forecasting is the only way.
4 Demand for capital goods follows the first derivative of demand for consumption goods
When analyzing product demand, a distinction should be made between consumption goods, like toys and clothes and capital goods, like machines and trucks. Capital goods can be defined as goods that are used to manufacture consumption goods. Demand for capital goods can again be split in demand for growth and demand for replacement. Demand for growth means the production capacity needs expansion. If the production capacity grows at a constant rate, growth-demand for capital goods is stable. If production capacity is constant, growth-demand of capital goods is zero. If there is overcapacity, growth-demand of capital goods is zero or negative. This effect is equivalent to the mathematical expression that growth-demand for capital goods is “the first derivative” of demand for consumption goods. This law may seem trivial or overly mathematical, but it has important implications. Growth demand for capital goods goes through a peak when the demand for consumer goods merely goes through its inflection point: thus much earlier in the economic cycle. For the same reason temp labor suppliers like Randstad always claim they are early-cyclic, as if they are leading. In fact they are not: they are just the first derivative of the main cycle. Another implication is in a country like China: China has grown strongly over the last 25 years, but mainly in capital goods for the own industry and export of consumer goods. When the export growth slows down (so it is still growing, only slower), Chinese production of capital goods will decline strongly.
5 Replacement of capital goods is installed capacity divided by its life time
We described that demand for capital goods can be split in demand for growth (which was discussed as law 4 in the previous newsletter) and demand for replacement. Demand for replacement means existing capacity is at the end of its practical life time. It is simply equal to the installed capacity, divided by the average life time. The installed base is equal to the cumulative inflow minus cumulative outflow. In most markets and business columns the total installed capacity of a good is known or can be estimated, as is the average life time. We like to refer to such installed base as a Fleet. Examples are the machinery fleet, truck fleet, and airline fleets. Also all non-residential construction can be considered a Fleet, and all office furniture and all office coffee dispensers and all printers. All mobile phones in the world are a fleet, as all ballpoints in use and the sum of all traffic lights. The replacement flow for these fleets of capital goods is more stable than the flow of buying capital goods for growth (which, remember, is the first derivative of underlying demand, according to our 4th law). There is one complication: under stress people sometimes decide to wait with replacement. This can result in what we call Aging Fleets which will be described in the next law.
6 Fleets maintain an optimal age
We defined a fleet as a stock of any equipment that is used for a longer period. We can now add that any Fleet not only has a certain size but also a certain average age and that is because this is optimal for a reason. The invisible hand of the free market makes sure the Fleet is maintained around that optimal. When a crisis hits us, the place of the optimum may temporarily change. But after the crisis, when things return to normal, the Fleet will return to its optimal size and age. Some examples: The famous Men’s Underwear Index says that men buy less underwear in a crisis; they let their boxers age. Flostock’s Sixt law says that this can only be temporary, and when the crisis is over, catch up demand will take place. Another example is the automotive fleet. When a population is uncertain about its economic future, it may decide to continue driving the old cars a little longer and the average age rises. For a while higher maintenance and lower status are less important than depreciation. Later the Fleet will need rejuvenation to get back to the original optimal average age.
7 Fleets buffer
We defined a fleet as a stock of any equipment that is used for a longer period, and that includes machinery, houses, vehicles, cloths, tires, phones, tooth brushes, etc. : anything that is used but not consumed. The 7th Law says that Fleets act as buffer between flows. In fact, Flostock believes that the buffering effect of Fleets is the main reason we have economic cycles. The problem is that most people don’t look at Fleets as stock, and not as buffer. Most people have a linear view of the world and consider fleets inert, that means not influencing the flows. One example is that the massive introduction of winter tires has maybe doubled the Fleet of tires in use, but increased the flow (=sales) of tires only temporary.
8 Inflow = Growthflow + Outflow
This law says that the inflow into a fleet or stock is the sum of what needs to be replaced, so the outflow, plus what is needed to adapt the fleet to the growth: the growthflow. “So what?”, you may ask.
This law explains car sales in the world, trucks, machinery, any fleet. Suppose capital goods need replacement every 20 years and the growth in demand is 5% per year. In such case, in equilibrium, Growthflow = Outflow. Another example: because global production went through a peak in 2008, the capacity now is still big enough for current demand and there is little growthflow in capital goods.
The 8th law is the integration of Flostock’s Laws 4 – 6 about Fleets. The Growthflow is based on the 4th Law. Outflow is based on the 5th Law. The 6th law explains a regulation mechanism of the Outflow. And the 7th Law warns that a change in Outflow can strongly influence the Inflow. Readers will understand that these effects together are too complicated to be understood by thinking alone. A good system dynamics model is needed to calculate the effects.
9 Stocks have a memory
Stock memory is like potential energy, ready to start flowing. A lake full of water for hydro-energy. A bucket of marbles balancing on a hill top. A syringe that is drawn vacuum. An under-stocked warehouse. A hill full of snow in spring. An empty shelf. A jungle full of pharmaceutically interesting plants. An empty terrace with a good view. DRAM memory. All savings on the bank. A pension fund. A group of hooligans in a bus. A full battery. Hot water in a boiler. So I use the word “stock” as in stock & flow, not just as a physical inventory. The flow from the stock gets direction from the equilibrium level: when snow melts it flows down. When people retire the pension fund sends them money. When the syringe is released, it will close. An empty shelf will be filled. Low inventory creates replenishment orders to fill a gap. In thermodynamics they call this equilibrium a “lower entropy level”.
In Flostock models this property is used for a flying start: a model can be started at any moment if the stocks (memories) are filled to the right level. This memory property makes it also possible to forecast non-linear: or, to say it in other words: the absence of stocks in a forecasting model makes it impossible to predict anything else but straight lines. This, we believe, is the main reason why most institutions have their forecasts wrong at every turn of the economy.
10 Commodities are locked in.
Meaning that there are very strong restrictions in capacity and storage, while demand is strongly dictated by the end market demand. On top of that, expansions in general need a very long lead time and are capital intense, so are only made after due consideration. Competition is hard, thus margins are at the minimum. This locked-in setting explains the high price elasticity of commodities up and down and it explains the often seen volatility. When a shock like the Lehman Wave takes place in these industries, it can take years before the water calms down again. And the lock-in explains what we call the commodity investment cycle. Fortunately, all these factors can be taken into account when building supply chain models for high volume products.
11 Lead-times make the world volatile
Lead-time, also known as delay, creates oscillations in all processes around the world, not just in supply chains and logistics, but in chemical installations, politics, societal changes, and guitars as well. Without delay or with a shorter delay most oscillations would not exist or dampen out quickly. And longer lead-times give more and bigger oscillations, so the of-shoring trend of the last 20 years has made the world much more susceptible to volatility. Your warehouses at the far end of the world probably have the biggest fluctuation in storage level. Demand for Made-to-Order products often fluctuates more than for Made-to-Stock products. Capacity for electricity generation is cyclic. Slowly but steadily growing smog (i.e. with long lead time) in China will kill its industrial growth before it recuperates. The monthly variation of your sales is also caused by lead time. 3D printing will reduce oscillation. Good planning, a reliable forecast and full transparency could prevent oscillation, but in many cases that is not possible and volatility is the result.
12 Price follows volume
Price follows volume, meaning that in a free market volume developments are the strongest driver in setting the price. Further we know that price elasticity is low, so if a supplier changes his price with a percentage x, the change in volume is lower than x and often much lower. In formula: Δ(demand)/Δ(price)<<1. That means, in the opposite reasoning, that if demand changes strongly, e.g. in a crisis or a panic, and supply is restricted, the prices may vary greatly, both up and down. Formula: Δ(price)>> Δ(demand). Flostock has developed a model for this. And since down is limited by the bottom (=variable cost, probably), the upward price peak is much higher than the downward price peak. Stockpiling and panic exacerbate the issue further. Confused? Call Flostock and we’ll explain what you can do with this in supply chain modeling.
13 All Oscillations are damped
All oscillations in the real world are damped because there is always friction and there is no such thing as perpetual motion.Even a gigantic oscillation like the orbiting moon is damped by the tide of the oceans: its circular speed is slowing down and at some point in time the moon will crash into the earth. The same is true for economic waves. Phenomena like the hog cycle, which seemed to continue forever, in reality are damped. The Lehman Wave was certainly damped, although much less than most people anticipated. Royal DSM and Flostock wrote in a recent article, that the damping of a single destocking pulse in the resins business was only 40% per year, so taking 4 years to subside.
14 Oscillations continue only if energy is added
We know that if you drop a golf ball on concrete it will bounce a few times, then lay still. Now suppose a golf ball is bouncing down a long downhill road with a stable slope and the extra energy from going downhill (mgΔh) is just enough to compensate for friction and elastic energy dissipation: the bouncing may continue for as long as the slope allows. When the angle of the road steepens, the bounces become higher, and can escalate to extremes. When the road levels off, the bounces become smaller and can disappear completely. Suppose in a second scenario that a ball is rolling, not bouncing, down the same downhill road, so the ball is not in oscillation. When the ball hits a small stone or crack in the surface it may start to oscillate and depending again on the slope, it may continue bouncing for a long time. The same applies to oscillations in the economy and the supply chains, including the hog cycle, and the IC industry.
15 Human behavior is stable
If we define behavior as “the response to an input”, we have found that, contrary to what most people think, most managers do not adapt their behavior to the circumstances. When someone’s behavior is split in small enough domains, it appears that in each domain the behavior is stable, also in a heavy crisis. The managers have targets, rules and regulations and continue to work according them. Moreover they have habits and routine. Some behavioral elements do follow the crisis pulse, but the response of each element is the same as outside the crisis.
So if normal purchaser’s behavior is to order the average of last month’s sales, this will be continued also if sales drops in a crisis. When the crisis hit the coating industry, and Flostock predicted an upward peak, all experts said that this would not take place because supply chain professionals would be “much wiser” and adapt their behavior. Nothing of the kind: the wave took place completely as predicted and everybody bought product in the same way as before, resulting in big overstocking. This continuity in behavior makes modeling and forecasting possible.
16 Impatient supply chains oscillate
Thanks to the work of Maxi Udenio described here, it has become crystal clear under which circumstances systems oscillate. This can apply to any system with stocks, flows and feedback loops with lead-time, so to many, many businesses in the world. In general, the stocks in such a system have a desired level, and the delta is called the gap. In impatient systems, people hurry to close an occurring gap.
XX deleted for proprietary reasons. Contact us if you wannt to know more XX
17 “Prisoner’s Dilemma” maintains market oscillation
Oscillation has debilitating effects on a business, so it is surprising that so many businesses accept it as an inevitable part of their existence. More than this, they actively contribute to it via their sourcing decisions. Flostock has identified a link between the well-known “Prisoner’s Dilemma” and the phenomenon of market oscillation. Understanding this connection can help companies mitigate the harmful effects of oscillation.
As stated in Flostock’s 16th Law of Demand, impatient systems oscillate. Stocks generally have a desired level. When that level is in danger of falling below acceptable limits, people hurry to close prospective gaps. The time for attempting to close such a gap is called the leveraging time.
XX rest deleted for proprietary reasons. Call us if you want to know more XX
A remarkably large proportion of business activities are in oscillation at any given time, driven by the fear of being unable to keep their customers supplied. The fact that various players are all eager to compete perpetuates the system: they are (naturally) keen to compete downstream by keeping their customers supplied, and so they order too fast upstream (from their suppliers).
If all players would relax and take their time over re-ordering stocks, the effects of oscillation would subside. If only one player were to do this, however, that player would be at a disadvantage compared with the others who are still ordering plentiful stocks. In Game Theory, this phenomenon is called the Prisoner’s Dilemma: all players would benefit if they would only collaborate, but the fear of being outdone by one another leads to behaviors that disadvantage them all. We call this Flostock’s 17th Law of Demand: “Prisoner’s Dilemma maintains market oscillation”.
Click here for more on the Prisoner’s Dilemma. For help identifying the causes of oscillation in your own market and developing strategies to deal with it, please email info@ or call us on +31(0) 6 11356703. flostock.com
18 Flostock Formula for Investments
Investment(t) = Demolishing(t) + d(underutilization(t))/dt + α*d(Sales(t)/dt +α* d(Inventory)/dt2
whereby α is a conversion factor describing how many inventory units can be made per capital unit per time, and ignoring delays. In words: Investment is equal to replacement plus a higher idleness plus growth in sales plus a change in the growth of inventory. This formula helps understanding the dimensions and the dynamics of investments. E.g. Sales and Inventory follow iron laws anddictate that investments are made, certainly for the industry as a group. If Sales goes up and nobody would invest, stocks deplete, a shortage develops, prices goes up and investing becomes more attractive, so will eventually be done. Delay in sales-driven investments creates wave around the α*sales curve. Lack of credit during the Lehman Crisis had two effects: 1) it caused companies to reduce their inventories strongly, making d(sales)/dt strongly negative, and 2) it prevented people from investing. Companies who could not afford a large d(underutilization)/dt, had to close, which is equal to strong demolishing. Surviving companies delayed demolishing, so delayed investment for replacement and let their fleet of capital goods grow older, but not bigger. This aging fleet will require extra investments in the future (see above Law 6). This formula works for a company, and for an industry, and for the whole economy, if the proper settings are used. One caveat: in reality delays play a prominent role in investments: so calculating real investments should be done dynamically.
19 Stocks give predictability
Most companies looking for a forecast want to forecast a flow, like sales. Hardly anybody is forecasting a stock . Nobody is forecasting inventory. The fact that the Flostock models can predict the future, however, is based on the effects of stocks. If Stocks are in play, and in real worlds that is always the case, the Flostock methods quickly demonstrate their added value. Stocks can be gigantic compared with flows, and when these stocks start moving and add to the flow, all normal flows If the trigger is known, the resulting flow can be predicted because most supply chain behavior is stable, so the speed of moving of the stocks is fixed.
You, dear reader, probably don’t have a clear view of a moving stock. You may think of melting snow, or the breaking of a dike, or the tides, which are extreme forms of shifting stocks. But seemingly stable stocks that shift a little bit can dramatically change the flows. Look at all the capital assets of the industry: when credit became short in the crisis, the stocks changed behavior: their owners decided to let them age, and as a result, the normal replacement flow dried up almost completely. Stable manager behavior and fixed supply chain structures make it possible to forecast the flows that are generated by this movement of stocks, like the strength of the tidal flow in the Channel is caused by the distance of moon and the shape of the coast line.
20 Stocks cannot be stopped
Flows can be stopped by people at any moment, immediately, and restarted immediately, that is, if the source stock and target stock allow it. Stocks, au contraire, cannot be changed easily and can certainly not be stopped. If the out- and inflows of a stock are stopped, the stock freezes but continues to exist. It never goes away. When you come back after a week, everything is still there. In contrast: if a flow is stopped, it immediately ceases to exist, for 100%. When your customer stops buying, your sales stops, but your inventory, your capacity and your debts remain. When the world stops burning fossil fuels, the CO2 flow stops immediately, but the amount of CO2 in the atmosphere will continue to be there and contribute to the greenhouse effect. This inertia of stocks is what makes them reliable, but also a hindrance. Just being there already cost money, in warehouses, occupied cash, interest, etc.
21 Gaps are proportional with adaptation time
Gaps, defined as the deltas between stocks and desired stocks, are proportional with adaptation time (AT); the period a manager allows himself to close the gap. In a continuous situation, with a gap that is never fully closed, the gap becomes proportional with AT. Intuitively, managers don’t want large gaps, so they order replenishments with low AT. If the AT they use is smaller than the lead-time, they may cause oscillation. If you don’t believe this, try ordering your goodies faster and observe whether your supplier will become more volatile. If you want to know more, give us a call.
22 Stable Fleets can have internal volatility
A building full of machinery, a fleet of trucks or a safety stock of canned food can superficially look stable, safe, and inert. But be aware: such a fleet can contain a hidden time bomb of volatility if the fleet build-up was not evenly spread over time. In such case the replacement will also be unevenly spread over time. Examples are the asset built-up of China over the last 20 years, the buying of a full wardrobe of fashionable clothes at the start of your career, and the aging of the baby boom workforce.
and next ...
The Flostock Laws of demand will continue to be expanded monthly with new insights around demand. As we see it now, the list will also include items around volatility, bubbles, monthly oscillation patterns, damping, and lead time. If you can't wait that long, call us now.