Will the AI dynamic pricing lead Hong Kong's shoppers to virtual insanity?

Price variance on basic supermarket items from week to week can be surprisingly large. I presume it's based on inventory, as some "deals" for various items can halve the price. It's one of the reasons I am wary of being a member of various rewards programmes and reluctant to hand over my full consumption profile - at a certain point they would get wise to the fact that Neil Newman will clear the next shelf of discount Cava at the weekend.

There has also always been a certain amount of dynamic, or rather arbitrary, pricing in wet markets. Without displayed prices, I get sized-up by the vendor wondering just how much I might be willing to pay for a bag of juicy lychees.

Do you have questions about the biggest topics and trends from around the world? Get the answers with SCMP Knowledge, our new platform of curated content with explainers, FAQs, analyses and infographics brought to you by our award-winning team.

It was apparently the Quakers that some 150 years ago in their belief of fairness and equality came up with the idea of a price tag, with storekeepers employing a fixed-price system. Other shopkeepers copied the fixed-pricing as they saw the benefit of not requiring their shop staff to determine prices on the fly. With the price of goods from distributors and manufacturers relatively stable, and staff not needing as much training, retail stores could get bigger, allowing the arrival of supermarkets and hypermarkets. It was good too for customers, who now knew where they were with their shopping basket.

Today, they have heaps of real-time data and computers galore to maximise sales and profits. Airlines may change the prices of their flights depending on the day of the week, time of the day and how many days there are before departure, even up to the day before departure. To do this, they use inputs that range from the availability of seats and departure time to the number of cancellations on other airlines, and, of course, that awful fuel-surcharge policy because of the volatile price of fuel.

Similarly, in hospitality we're now used to the time-based prices of accommodation, with peak prices during summer and school holidays, for example, and lower prices when fewer people are travelling. At all times, like my lychees, the game is to figure out the highest amount consumers are willing to pay.

So this works well for retailers, but what about manufacturers? Do they set prices dynamically? It appears less common.

I know a small manufacturing firm, Commtel Ltd, which makes 'Telguard' intercom terminals. They agonise over price increases and go to some lengths to hold prices stable so that their clients - distributors and installers - can plan their businesses, inventory levels and profit margins when selling to retail consumers. Commtel is primarily exposed to the prices of electronic components now in short supply, stainless steel (up 76 per cent in a year), aluminium (up 68 per cent), and unpredictable surcharges for shipping. So for Commtel, the annual event of calculating a fair price increase to maintain margins is stressful, time-consuming and risky.

If Commtel's process for reviewing its output prices is representative of other manufacturers, a whole lot of number-crunchers in offices around the world must be having a very bad year. Input prices of raw materials and commodities have soared through 2020 and so far this year, while the cost of transport has also sharply increased. And don't get me started on oil, where we've gone from not being able to give it away, and a negative price of minus US$37 per barrel at the April 2020 futures expiry, to as much as US$74 per barrel currently.

Yet it appears the sharply increased input costs are not filtering through into official inflation numbers. For example, the small uptick in Japan's Consumer Price Index (CPI) in May, a puny 0.1 per cent increase and the first time it went up since March 2020, created something of a media sensation. Similarly, in China the official numbers suggest the same, with the May 2021 CPI up just 1.3 per cent, even as its Producers Price Index shot up 9 per cent year on year, showing the biggest rise in inflation at the producer level in 12 years. Still, this seems like a tiny increase compared with skyrocketing input prices for the likes of iron ore (up 67.2 per cent), plastics (21 per cent), soy (12.8 per cent) and of course copper which posted an increase of 105 per cent in 12 months.

Given the difficulty businesses now face in holding goods prices stable for the retail market, I was wondering if dynamic pricing at retail level may become further driven by dynamic pricing at the distributor level, the manufacturer level and possibly even the raw material sourcing level. The knock-on effect of this would be to influence retail pricing on a daily basis in a new way.

In February Tesla had offered discounts, but in April it hiked prices for several car models, twice. And at the beginning of May, it again raised prices for various Model 3 formats, and the Model Y, by US$500. Tesla has a history of keeping its prices volatile, but recent rounds of price changes have been very rapid.

Readers may think I am taking a rather large leap from car manufacturing to high street shopping, but input prices have risen for more than just microchips, metals and ores. Take for example chicken (up 76 per cent in a year), pork (55 per cent), corn (a whopping 142 per cent), coffee (56 per cent) and cotton (62 per cent). Does it not make sense that the price management process that works for retail should apply to manufacturing and manufacturing inputs? That means dynamic pricing evolves into a hellscape.

US manufacturers are largely still using manual tools and warm bodies to revise output prices in the wake of volatile input prices. But people are reactive, relying on instinct and experience, and in a volatile pricing environment their instincts can be wrong. More and more manufacturers are therefore looking closely at switching to dynamic pricing algorithms.

The 2021 B2B Distribution Pricing Survey of thousands of distributors by epaCUBE revealed that pricing is the number one pain point for global manufacturers, and is reportedly resulting in an annual margin loss of between 0.51 per cent and 7.47 per cent.

Consultants and vendors of complex AI systems are already hard at it, selling their services to manufacturers as well as retailers which will in turn develop their own algorithms to outfox the consumer and competition on pricing.

We could get to the point where both online and bricks-and-mortar retailers will change prices rapidly, just as the airlines do, in response to upstream factors, not just storefront traffic and inventories, to protect their margins.

At that point us consumers will be driven insane, as we would have no idea what anything may cost in any store at any given time on any given day.

Neil Newman is a thematic portfolio strategist focused on pan-Asian equity markets

This article originally appeared on the South China Morning Post (SCMP).

Copyright (c) 2021. South China Morning Post Publishers Ltd. All rights reserved.