Academics, Consultants and Practioners

In my backlog of RSS feeds, I came across an article titled Pulling Cars and Allocations on how poor application of pull systems and allocation rules for the reduced production have led to trouble in the car retail business. (Essentially, dealers can't get the cars they want.) The articles notes that this problem arises due to: (i) Car sales happening daily but allocations happening on a longer time scale — usually monthly, and (ii) variety not being handled properly, and (iii) the stickiness of allocations. I'll talk briefly about (ii) since the article explains (i) and (iii) in sufficient detail.


Operations Management folk (and those who took MBA classes on operations management) would generally love pull systems. The fact is, a properly implemented pull system that is in harmony with the conduct of retail operations greatly reduces inventory in the supply chain, meaning that less product is just sitting in some warehouse or storeroom. Pull is wonderful for goods that are standard or of limited customization.


However, where retail operations are such that one does not sells "cars" but an assortment of models and makes with various major options. The fact is, each customer comes in with demand for one car and, implicitly, a short list of variations he/she would like/accept in some preference order (and with a range of acceptable prices). This implicit list of acceptable cars can be very short or very long (for the less picky). Now if a customer comes by with a short list of acceptable options and can't find one on the, now sparse, lot, there is a substantial likelihood that the sale would be lost. It would appear that retail operations here behaves like maintenance operations for systems that require a high up-time — the particular spare parts and the particular maintenance specialist must be on hand at the time.


Since in most business transformations, consultants are called in to make concrete recommendations on what to change and how, I blame bad consultants for this mess. Clearly the end-to-end operations were not properly considered in the implementation of the sexy new "lean" paradigm. The fact is, this is not a case where "we didn't expect this to happen" could be responsibly argued.


Before laughing their way to the bank, the consultants could, at the very least, have done some thought experiments on what would happen if a customer came in and happened to not find what he/she wanted. Tracing that to the allocation system would have given some reason to make the system more robust. Moreover, such a scenario would not be of the "Black Swan" variety, necessitating some analysis on this particular "use case" of the "car manufacturing and distribution system". (In fact it would be more like the random variable "Is there a swan in Swan Lake in the time interval [T₁, T₂]?") Unintended consequences can surely not be argued in this case.


Some time ago, I read The Boston Consulting Group On Strategy (2nd Ed, Wiley, 2006, edited by Stern, C.W. and Deimler, M.) wherein Bruce D. Henderson, the founder of the Boston Consulting Group, (and the man most responsible for the famous BCG 2x2 growth-share matrix) was said to believe that:

... while most people understand first-order effects, few deal well with second-and third-order effects. Unfortunately, virtually everything interesting in business lies in fourth-order effects and beyond.

I've never observed a major fourth-order effect. At least not in a non-historical, non-geological, non-evolutionary or non-cosmological scale. Perhaps I'm just not perceptive enough. Second/third-order effects are things we all see. The worst thing about them are that they're like a one-way mirror: they're opaque one way, and you can see and understand their causes so perfectly on hindsight. On this note, I'd like to digress to talk about a well studied set of "unintended consequences" that were totally avoidable.

Consider the aftermath of the recent financial crisis of 2007-2009, everything looks perfect on hindsight. But prior to that money gots in the way of "the good ideas winning out". After the damage was done and there was no more money to cloud judgment, everything became clear and we wondered why no one saw things coming.


In Aug 2005, Raghuram Rajan, Chief Economist of the International Monetary Fund from 2003 to 2006, gave a speech about (i) how "competition" "forces" bankers and traders to "flirt continuously with the limits of illiquidity", (ii) how investment managers returns have a huge personal upside for taking on more risk and the downside is trivial, (iii) regulations should be in place to align incentives (this being my summary of his recommendations. As BusinessWeek (Feb 13-Feb 20, 2011) put it, "he was treated like a skunk at the party". It even noted that Lawrence Summers, former US Treasury Secretary and then president of Harvard, said he found "the basic, slightly Luddite premise" of Rajan's presentation to be "largely misguided". What people say is aligned with money. I find it absurd that an economist of Summers' standing could snort so pretentiously at the idea of aligning the behavior of agents to systemic objectives through the appropriate incentives.

Returning to original topic (after that rant of sorts), consultants to business play an important role of filtering academic output and tailoring it to particular business settings. Where they do, the results can be spectacularly good. Where they get lazy and transplant ideas without sufficient adaptation to particulars of an industry, unintended consequences arise. And these unintended consequences can show up after some time has huge negative effects. Laziness often takes the form of assumptions that explain away the need for more analysis (usually on incentives) or work (building institutions that align behavior to systemic objectives). Laziness explained away and still explains away the moral hazard of the financial industry with the now popularly rejected notion that bankers and traders are upright custodians of investors'/savers' money.

Consultants have an important role to play. While much new knowledge comes from academia, it is often general and abstract. Solutions arising from academia often take the form "if operations are structured like A and B holds, then C happens". Where the conditions of B are at variance with reality, it should be studied whether the outcome deviates significantly from the desired C. The value of consultants is their wider exposure to industry. Their recommendations should be made in the context of actual, real world, operations.

Truthful Voting through Approval Voting

Noting that enthusiasm is building up for the coming General Elections, it is timely to assess our current voting system. We use a simple plurality system where each voter may choose only one candidate and the candidate with the most votes wins. In situations with more than two candidates, this system entails the possibility of voters being in a situation where they have the incentive to misrepresent their preferences.

Consider a situation where candidates A, B, and C have 40%, 35% and 25% of the electorate in support respectively. B's supporters do not mind C, but do not want A to win. C's supporters do not mind B, but also do not want A to win. In the plurality voting system, C's supporters have the incentive to misrepresent their preferences and vote for B rather than vote truthfully and have A come into power.

It can be proven mathematically that where voters' preferences are represented by a ranked order of candidates, any voting system offers situations where voters have the incentive to misrepresent their preferences. This is an unfortunate and inescapable reality. However, where voters' preferences are represented as "Approve"/"Do not approve" ratings for each candidate, a voting system known as “Approval Voting” gives voters the incentives to vote "truthfully". In Approval Voting, voters cast "Approve"/"Do not approve" votes for each and every candidate, and the candidate with the highest approval level wins.

The truthfulness of the Approval Voting system rests on the idealized "Approve"/"Do not approve" representation of preferences. I argue that elections are not popularity contests but polls of the electorate to see who has a mandate to represent them. As such, a voter might conceivably support multiple candidates or none at all. With Approval Voting, the results of the election unambiguously describe "the mandate of the people". A winner with 90% approval can be said to truly have 90% approval from the electorate as there is no incentive to misrepresent preferences. Conversely, a winner with just 20% approval can be unambiguously seen to have a weak mandate.

In addition, this ameliorates the situation for the opposition with regards to election deposits, since a viable candidate may stand and have approval from a large fraction of the electorate, yet that same large fraction may prefer another candidate.

I suggest that such a system be considered for subsequent elections. This would be a worthy matter to debate after the upcoming elections.

Maintain a Surplus of HDB Flats

Some years ago, there was a surplus of HDB flats that was described in some quarters as mismanagement on the part of HDB arising from an over-projection of the demand for flats. I personally do not see having a surplus as a problem. A surplus of HDB flats enables first-timers to get flats sooner, perhaps a year or two earlier.

The government might like to think of the cost of empty flats as a subsidy in terms of time to first-timer flat applicants. Early acquisition of a home to call one’s own can make a positive difference to the willingness to have a children. As such, surplus flats are compatible with Singapore’s needs. The question then arises of how to manage the flat supply.

The quantity of the stock of flats should me managed so as to ensure a healthy but not excessive level of surplus flats. To do this, the key is to launch the next project only when the stock of remaining flats falls below some critical level. This critical level should be determined based on (i) the lead time from launch to the handover of keys, (ii) the number of flats to be built in the next project in the pipeline, and (iii) the estimated demand for flats over the next few years.

To manage the quality of the stock of flats, the key is to ensure that older flats get sold first by policy means (e.g.: not allowing a flat to be sold if a flat two years older is still on the market) or market means (e.g.: discounting older flats slightly).

Surplus flats can be a strategic tool to achieve national objectives, and should be used as such.