You can read Part 1 of this series here.
Let me get straight to the point. The Bullwhip Effect primarily happens due to two reasons:
1. Conflicting incentives
2. Lack of communication across the supply chain (information asymmetry)
Both of these reasons are founded in the primary axiom of economics and free markets, without which economics as a field wouldn't exist:
Self-interest; every actor in a market is assumed to act rationally, based on self-interest.
As we go through the examples, I want you to see this common theme emerging.
Let me start with an example that's most recent: the supply chain crisis during the Covid-19 pandemic.
Particularly, the global semiconductor chip shortage that ensued as a part of this crisis.
So, what essentially happened was: with remote work, consumer spending in many areas took a major nosedive. And all this expendable income found its expression in an increased demand for all kinds of work-from-home devices, consumer electronics related to the bazillion podcasts started during this time, and essentially all work/hobby electronics that promised some respite from the boredom of the lockdown.
As a result, manufacturers of these electronic products saw a huge surge in demand. And like any rational actor, these companies wanted to rise up to the occasion and service this demand. Because why lose out on business? Make hay while the sun shines.
But all this electronic equipment runs on semiconductor chips. So, all manufacturers — right from those who built smartphones, automobiles, and gaming consoles, to cameras, graphic cards, and other consumer electronics — started placing large orders to the few large semiconductor fabrication facilities that produce all of the world's computer chips.
This inevitably created a sudden surge of demand for suppliers like Huawei, Qualcomm, and NVIDIA, which design and sell these chips that go into practically everything.
These companies then put even more orders up the supply chain, to chip manufacturers like Texas Semiconductors, Samsung, and Intel which have chip fabrication facilities.
In turn, this led to even more orders for companies that supply parts to these chip fabrication facilities.
The impact spread like a shockwave across the market and ended up impacting companies that had little to do with semiconductors but still used electronic equipment.
We are in 2023 today, and many of these purchase orders are still not serviced. I can tell because a lot of electronic gadgets are still unavailable on Amazon. There are simply no chips to build them, I guess.
And what happens when this huge backlog of orders gets fulfilled? Manufacturers will be left holding a lot of raw inventory for products whose demand has now subsided.
This is also why the best part is no part.
Anyway, let me outline the primary factors that lead to the bullwhip effect.
1. Demand forecast updating
When operating at scale, businesses forecast demand and restock inventory accordingly.
As a sourcing manager, when you place an order with the supplier, the supplier sees that order as a signal about future product demand. Based on this, they update their demand forecast and factor it in while placing orders to suppliers up the supply chain.
And generally, to protect against volatility, as a sourcing manager, you will order slightly more than the amount you need to strictly replenish your inventory. And just like you're a rational actor, everyone up the supply chain is. So even they order slightly more than what they strictly need to meet current orders.
This is why the effect compounds down the chain.
You need 10. To be safe, you order 12.
Your supplier thinks you need 12. To be safe, they order 14.
Their supplier thinks 14 is the market demand. So they order 18.
And so it ripples across the chain.
This is why the signal the supplier at the end of the chain receives in terms of orders is starkly different from market demand. Consequently, they also face more demand variability than distributors and retailers.
To add to it, once you have long WIP and lead times for products, the variability increases manifold and creates more WIP.
2. Order batching
As we discussed in our batch vs. flow processing piece, batching is rational when there are some advantages like economies of scale to be had.
There are substantial differences between full truck-load and less-than-truckload rates, so companies have a strong incentive to fill a truckload when they order materials from a supplier. They get much better pricing, so they choose to restock infrequently (say once a quarter vs. once a week) and place larger orders.
But large orders lead to long order cycles. And long order cycles simply magnify volatility for the supplier.
For instance, consider a company that orders on a monthly basis.
For the supplier, there is a spike in demand at one time during the month, followed by no demand for the rest of the month. And you're just one manufacturer placing a large order. Like you, there are several others.
So what happens when, due to market conditions, a supplier suddenly receives large orders from multiple companies down the chain at once?
Infrastructure gets overloaded. You get much longer lead times than usual. There are backlogs.
And you know what's worse?
If as someone who placed an order and received a lead time of 1.5 years (which isn't unusual in the semiconductor industry), market demand might change in the next 8-12 months and you will cancel your order.
Like you, many others will, too. Your supplier is now left holding a bag of excess chips no one wants.
3. Price fluctuation
When something is available for cheap, any self-interested rational actor will buy more than what they need, simply because they're getting it for cheaper.
This is called forward buying.
A lot of buying in a supply chain is forward buying: manufacturers buy inventory way in advance of actual consumer demand, simply because the supplier has offered a price discount.
Now that you've bought more than you needed, you will not buy again until you deplete your inventory. As a result, your supplier faces huge bumps and dips in demand. Their factories are running extra time during certain times and totally idle at other times.
Your supplier will in turn build up huge reserves of inventory to buffer against this volatility in demand.
And inventory storage is expensive.
The cost of maintaining said inventory without spoilage is also expensive.
The irony? This expense will eventually get reflected in the price you're paying for your products.
So, for want of a cheaper product, you (and other actors like you acting in self-interest) just increased its price in the long run.
This is called "playing yourself." As in,
"Congratulations, you played yourself."
4. Rationing and gaming supply
If a supplier knows that you're ordering in excess of what you need and their production is limited, they will ration their supply to you. For instance, you may have asked for 10,000 pieces, but they will tell you that we will only service 70% of your demand right now.
But you're one smart aleck and you knew this was coming. So, instead of ordering 10,000, you will order 15,000 so that you get 10,000.
Unfortunately, what this does is it sends up a completely skewed impression of real consumer demand for a product up the supply chain.
To add insult to injury and exacerbate this skewed signal, self-interested players will place the same order with multiple suppliers and buy from the first one that can deliver, then cancel all other orders!
You wanted 10,000 pieces of something.
You ordered 15,000 pieces of that thing from four different suppliers.
You got your 10,000 pieces from the supplier who delivered the fastest.
But up the supply chain, production was ramped up to produce 15,000 x 4 = 60,000 pieces.
Now, that remaining excess inventory will sit with your supplier as demand tanks and the market is overwhelmed with the ordered goods, and in the long run, this will only lead to the whole world getting products for a lot higher cost than they would have paid if some measures to avoid this were taken.
As to what those measures are, we will discuss them in Part 3.
Till then, I want you to think about the Japanese Keiretsu System and how it can mitigate this effect.
