Since last year, we have more and more bad news in the business sector. Various famous or long-standing shops shut down, triggering more and more worries about the local economy. One notable voice associated with these closures is that landlords refused to revise down the rents for shop spaces in spite of the worsening market situation. As such, these retail operators, given declining revenue, were forced to shut down their businesses.
Do all these make sense in economics? At its surface, there is nothing wrong. The original retailers facing a declining business of course want lower rents, which is the only way for them to rescue their businesses. But the landlords do not need to share these retailers' concerns. The shop spaces they owned can be rented to any other retailers. If the original retailers can't secure sufficient businesses to pay a high rent, perhaps others can. There is no reason to revise the rents down unless all other retailers also can't pay a high rent. But then let's wait and see. If these landlords can't rent out their shop spaces for a long period of time, they will eventually revise the rents down.
Hence, everything appear to make sense. But things are not that simple. The process of adjusting rents down is said to be too slow. What we can observe in the market is that rents do adjust down. But some people believe that existing adjustments can't reflect the fundamentals. Vacancy rate in the retail shops is high enough for a sufficiently long period of time. If landlords simply want to wait and see, they do really wait too long. Their belated response can't be justified by information lag. These are some viewpoints aired in the market.
Should we take this viewpoint seriously? Should we say adjustments are too slow? In a way, this is a subjective judgment. There is no reliable criterion for us to judge if the market adjusts too slowly or appropriately.
Someone says the retail sale data has declined much faster than rents. This is an indication that rents decline too slowly. Nonetheless, retail sale and rent can't be compared so directly. In fact, when retail sale increases, normally rent also won't increase as fast. Of course, rent may decline much much slower than retail sale. When the degrees of change differ very much for the two variables, up to a point we tend to think there is something wrong. But how much slower is too slow? There is still a subjective judgment involved. For economists, normally they won't make such kind of subjective judgments. Furthermore, they normally will (perhaps blindly) believe in market efficiency. They won't blame the market as adjusting too slowly.
But anyway, market players are sufficiently serious about this judgment. Some of them actually raised a hypothesis to explain why the adjustment is slow.
Here is how this hypothesis says.
Many landlords did not self-finance their purchases of their retail properties. Instead, they borrowed money from banks to finance their purchases. Banks made loans based on their valuations of the properties. For example, for a property valued at $10 million, a bank may lend 50% of the value, or $5 million, to the landlord. However, if the property value fell (as what happened these years), say, to be close to or lower than $5 million, the bank will require the borrower to return the money immediately. The landlord may need to sell the property immediately to cope with the bank's demand. The landlord may think the property value can recover to $6 to $7 million in a year but the bank's call disables the landlord to wait and avoid the value loss.
The story, as told, is this. Since these landlords want to gamble on market recovery, they don't want banks to immediately call back the loans made to them. To facilitate this, they would rather leave their properties vacant instead of leasing them at a lower rent. This is because banks' valuations on properties are based on rental income generated from properties.
For example, if a property's rental income is $5 million, and a bank believes that a reasonable rental return is 5%, then it will value the property at $100 million ($5 million/5%). If the property is now rented at a much lower level, say, $3.5 million, the bank will value it at $70 million ($3.5 million/5%).
Fearing that bank's valuation is sharply revised down, the landlord may leave the property vacant. Doing so, the bank will not have sufficient data to assess the rental income and may wait for a while until more data can be collected. As such, the landlord may be able to buy time for market recovery. Nonetheless, when many landlords do this, the market adjustment in rents will be slow.
This hypothesis has been aired in the market for some time. I have encountered it at least twice in some talk shows. In one time, it has been mentioned by a property agent (called Agent A) specializing in retail properties. In another time, even an economist mentioned it while in the same occasion another property agent (called Agent B) not specializing in retail properties didn't believe in this hypothesis, saying that bankers wouldn't be so stupid.
Interestingly, in this second occasion, the economist's viewpoint looks more like a property agent while the property agent's looks more like an economist's.
Now, turn to my original question: Does all these make sense in economics? In particular, does the hypothesis of bank valuation make sense?
As mentioned before, economists have a tendency to believe in market efficiency. If so, they should tend to reject the above hypothesis (like Agent B above). Why? The point is exactly that we don't have an objective method to judge if the market adjusts too slowly or not. Without such a method, economists tend to simply take what happens in the market as what is most appropriate. Thus, if the market adjusts slowly, that should also be the best adjustment speed. The market is efficient.
As an economist, do I take this view? Honestly, I don't. I think taking what happens in the market as best is simply a lazy way of thinking. Also, it is not meaningful to take this line. It tells us basically nothing about the market, except that it places a colourful word "best" on it.
As an economist, we should be more interested in understanding how the market works and how could we change its direction (so that it comes closer to our wanted situation). From this perspective, the naive market efficiency doctrine as described above is not a useful and good doctrine to be followed. Though some economists embrace it, not all will.
In this sense, Agent B above thinks more in line with this naive efficiency doctrine. He can't believe market players, notably bankers, are so stupid. So, the slow adjustment can only be interpreted as these market players truly can't discover the real market situation and therefore adjust slowly. In a sense, Agent B thinks like an amateur economist. But I don't think his belief (say, bankers can't be so stupid) is truly substantiated. He simply believes in the market.
In this sense, the economist mentioned above, believing in the hypothesis, is at least not so naive. He is not like an amateur economist who says what happens in the market must be the best. At least he has tried to uncover a mechanism that can explain why adjustments would be slow. Though the hypothesis may be wrong, it is at least a description about how things work.
So, do all these make sense? My point is twofold. For one thing, I don't think blindly believing in the market is a sensible attitude. Under this attitude, everything makes sense. But it is not sensible to believe that everything makes sense. What we need is a hypothesis that has potentials to explain what happens.
Then, eventually we have to assess if the hypothesis about banks make sense.
As mentioned above, I heard the hypothesis twice and the first occasion involves also a property agent, Agent A. Unlike Agent B, Agent A specializes in retail properties. Agent B doesn't specialize in retail properties while Agent A does. Thus, it is more reasonable to expect Agent A knows more about retail property owners' concerns than Agent B. Then, if Agent A raised the hypothesis above, it is less likely that the hypothesis is completely artificial. Perhaps he misunderstood some details. But he has no reason to fabricate the entire story.
So, what appears to be true is that landlords do have a concern that lowering rents fast will fasten bankers' downward adjustments in valuation, which eventually triggers loan recalls. The tricky point is: Why bankers also wait together with the landlords when rents stand firm but vacancies increase? Why bankers didn't take the initiative to re-value these retail properties, and took quicker actions to recall loans?
In fact, I think a banker will take fast actions if they observe other bankers also do so. If a banker calls back a loan from a borrower, the borrower will have a smaller capital and this makes it less able to return the money to other bankers who have also lent money to the same client. To avoid this, a banker would want to be the first one to call loans instead of the second. In such a situation, bankers have pressure to take fast actions.
However, bankers do not have such a pressure at the moment. There is no such a wave of calling loans at the moment. If you move, then I will move. But you don't move, and so I don't need to move. This is what happens at the moment. Hence, what we need to explain is why no one moves at the moment. Why do bankers not take the initiative to re-value properties and call back the loans?
One simple explanation is bureaucracy. A bank is not a person. It involves a hierarchy. Decisions may be made in steps. Thus, slow motion is the norm, not exception.
I do not know how far this can explain the phenomenon. But I think there is a limitation of this line of explanation. It must be that the decline in property value is not too obvious. Otherwise, despite bureaucratic, bankers must still take actions to protect their own interests. Therefore, to me, the key is why the value depreciation is not obvious, though some outsiders think that it is obvious.
At this point, I think the answer is almost found. The answer is that the degree of value deprecation is truly not very obvious. We may think it is obvious that the property value should have depreciated a lot. But to each specific property, how much its value has declined is not so obvious. The key is that retail properties are extremely heterogenous. Each shop differs from another a lot and so each shop's value can't be easily assessed.
In a way, we may say all properties are heterogenous. A residential flat on 10/F of a building must differ from the flat next to it on the floor above (or even on the same floor). Nonetheless, the difference is not so great and so there can be an easier way to assess how much different the values of two flats should be. Flat A may differ from Flat B as the former has 100% sea-view while the latter has only 96%. Such a difference is calculable and can be properly valued.
The value of a retail property is not determined in a similar way, however, as its value depends crucially on the business revenue that can be generated from the shop. Of course, location is an important factor. But rarely do we observe that next to a coffee shop is also a coffee shop though it may be a restaurant. The reason should be clear: a new coffee shop owner doesn't want to open a shop so close to an existing coffee shop such that potential clients are spread. So, you can see: location may sometimes be just a secondary factor in determining the value of a retail property.
In fact, the situation is even more complicated and the logic is not strict. Unlike coffee shops, sometimes we find bubble tea shops are clustered together on the same street or even next to each other. Why don't they fear that each shop steals business from each other?
On the one hand, clients come and go quickly for bubble teas. Seats are not offered while the consumers also don't need seats. As such, each bubble tea shop will not occupy their clients for long. The clients actually flow in from everywhere nearby. A coffee shop will fear that their clients will be occupied by another shop next to them. But bubble tea shops won't fear because the former type of shops occupies the limited number of clients for quite a while but the second type won't.
On the other hand, clustering may generate a promotion effect that attracts clients wanting bubble teas to come to a specific area, which bring more clients than without clustering.
Now, you can see: the values of retail properties depend on many factors and can't be easily assessed. It is much more complicated to evaluate a retail property than residential property. Yes, bankers may have some standard valuation models but they should also know that such models are not very reliable for retail properties. Shop A and shop B may be both located on the same street but their rents could differ greatly due to some factors that can't be easily captured in a standard valuation model.
In short, the heterogeneity in retail properties is not comparable with that in, say, residential properties. In this situation, rushing to revise down the value of a shop may not be a wise decision. Of course, when a property is rented at a much lower level, the bank will have sufficient data to revise down the value of a property. Their next action may be to call back the loan from the landlords, who don't want this to happen. Hence, this may create an incentive for landlords to stand firm on rent and wait for a longer while. Of course, while landlords may delay renting out a property, they can't wait for too long. Sooner or later landlords have to face the reality. It is simply that adjustment may be slowly made, and not as quickly as some people may expect.
