Toilet Paper in a Pandemic

This probably isn’t the best use of my time right now,  but maybe there’s a teaching moment here.

In the face of the COVID19 pandemic, folks in the US (and apparently most developed nations) have gone on a toilet paper buying spree.  One result besides the appearance of empty shelves, has been a lot nasty commentary in social media and news media attacking people for alleged “hoarding” of TP. The comments have come complete with  all the cheap puns and childish humor talk of toilet paper invites. (I will probably fall prey to that too!)

In store after store, the shelves often appear empty and denuded of toilet paper. This is a rather sudden and unexpected development for most people accustomed to a first-world life where a fundamental fact of like is assumed to be: there will always be toilet paper.  People and pundits have proclaimed a shortage of toilet paper brought on by irrational “idiots”. Social judgements and pejoratives have been flying since.

Even economists are not exempt. Justin Wolfers has likened the toilet paper shortage to a classic run-on-the-bank (couldn’t resist the pun, eh, Justin?) which might be individually rational but becomes a systemic failure. Justin considered the possible need for a national strategic toilet paper reserve.

I’ve read several pyschologists speculate on the “meaning” and “underlying psychological needs” driving the “hoarding” of toilet paper. I’ve read that it’s a controllable first-world comfort when people are faced with a sudden, scary uncertainty. Maybe there’s a tiny bit of truth in that, but it’s not enough to IMO to move that much TP. Other psychologists have speculated that it’s herd behavior. Some people bought up some more TP than normal, so they did too. Soon the shelves were bare and panic has ensued. Again, maybe a little sliver of truth, but not enough.

I’m going to weigh in on the topic, too. I think I have a bit more experience with toilet paper than the average person and more so than the average economist. It’s not because of my diet. Not only am I an economist and things like “supply” and “demand” are my basic tools, but early in my career I was the business planner/strategist for a very large industrial distributor/wholesaler. A distributor that distributed – you guessed it – toilet paper.  A lot of it. We didn’t supply much to the retail channel. We supplied businesses, factories, restaurants, hospitals, schools, and hotels.  Thanks to us, the nether regions of millions of people from GM factories to Florida resorts were kept clean and hygienic. I learned a lot about the economics of toilet paper back then. I’m sure it’s changed some since, given the better inventory management tools available today, but the essence of analysis still holds.

I think what we’ve seen has been a fairly rational response, at least initially, and I don’t see any great toilet paper crisis of 2020.  Yes, there has been some individual hoarding behavior, but probably a lot less than people think. What I see is a sudden shock to one of the most incredibly stable supply chains around. You see, there really isn’t any toilet paper inventory as you think of it.  Let me explain some of the economics of getting dried squares of wood pulp from a mill to your butt without discomfort.

TP-nomics: More than you ever wanted to know

Consumption drives the demand toilet paper.  Even in a crisis, there is not really any speculative value to toilet paper, nor is there any value to just owning a bunch of it. There is no ongoing use value either. Use two squares, and it’s done. It’s not like paintings which give ongoing value just for having them. It’s not a good store of value like cash, or gold, or bonds. It’s not a capital asset like a house, furniture, or a car.  It’s a single use value.  True, the perceived use value is high (at least in non-bidet using developed countries), but it’s very utilitarian. It doesn’t even bring social status, despite all the Charmin advertising.

I can tell you this from having analyzed and modeled the consumption and demand for toilet paper all those many years ago. It’s stable. Rock stable. Easy to forecast. If you know how many butts you’re dealing with, you can predict the consumption. That’s it. (and no, local cuisine or tastes don’t matter). People x days = TP needed.

The second fact of TP economics is that there’s a giant distributional mismatch. It’s produced in vast quantities in a very small number of places (mills) but consumed in a mind-boggling number of very specific locations. There are two in my house alone. When the need for it arises, the inventory MUST be in each location. When I want TP in the downstairs bath, I’m going to be really annoyed if it’s not there and I have to go source it from the upstairs bath or the basement pantry. I’ll be even more annoyed and greatly discomfited if I have to go to the store before I can use the toilet. Know what I mean?

In contrast, toilet paper is produced in a small number of places called mills in very, very large quantities on very, very large machines. So while we might indeed know how much TP is used each day – and that’s pretty much how much TP will be produced each day – we still have a problem. We produce just enough each day, but it’s in giant quantities in a few places when what we really want is a few squares in billions of places.

The third fact about TP economics, is that despite whatever its use value is, the dollar value per cubic foot of space it occupies is very low. It’s bulky. That means for a store or distributor it takes up a LOT of shelf space but doesn’t generate that much sales revenue or profit margin compared to other higher-value, smaller items.  At home, you only keep a roll or three in the bathroom. Nobody builds a bathroom cabinet to accommodate the giant Costco carton. For the store, the only way to survive those economics is to turn-over the inventory rapidly. That means that relative to demand, there’s actually very little in inventory. The large bulk makes it look plentiful to the consumer, but it’s not really. It’s simply being restocked, re-ordered, and replenished faster than just about any item in the store.

So how does our vast national TP distribution system work? It’s a network of very high turnover inventory locations, each one serving to “break bulk” in the distribution from mill to your butt.  The mill produces today and puts it on trucks or rail cars immediately and ships it out.  As soon as the truck/trains can get to the distribution centers (like the ones my ex-employer owned), it’s unloaded. A train car load is split up and becomes stacks of pallets or stacks of cases but they don’t sit still. That inventory is turned over very fast -meaning the cartons are loaded onto trucks and delivered to stores, factories, hospitals, and smaller wholesalers. Even back in my day 40 years ago, we emptied and shipped out that whole rail car in less than week. I wouldn’t be surprised if these days, turnover in some distribution centers could be measured in hours.

There really isn’t any toilet paper buffer inventory anywhere. As a nation, we haven’t really had a reserve of TP to buffer fluctuations in demand or supply – because we don’t need it.  Forests keep making wood fiber at about the same rate every day. People keep wiping their butts at the same rate everyday. What appears to be inventory is only temporary spots of relocated stuff that we turnover as fast as a possible. We have just-in-time production.

To recap:
Made in mill by train load –> whse in cases/pallets –.>store in multipacks –> home by carton–> bathroom cabinet by the extra 1-2 rolls –>dispenser roll

As soon as the minimum viable quantity runs out at each stage, we replenish. It’s a TP flow, not a TP inventory.  In reality, the slowest moving inventory is at the household.  Most households tend to buy one to a few weeks worth of TP at a time. They don’t want to waste space storing more. Assuming they don’t wait till they totally run out, that means they may have, on average, let’s say 10 days worth of TP on hand. When it gets low, they buy more. The grocery or big box where they shop knows that thousands of customers are doing the same predictable thing and the store only stocks enough for maybe 2-3 days worth of sales in inventory, maybe even less, knowing they’ll get more shipments.  The shelves always look nicely stocked with TP, but it’s all fresh, new stuff – unlike that jar of pickles you bought that was actually produced a year ago and been sitting in the store ever since.

So What Happened?

Everything is fine, until the Centers for Disease Control or other public health authorities try to gently prod people to “prepare” for possible isolation and quarantine because of COVID19.  It’s rational and it’s needed. The point of the isolation, social distancing, and quarantines is to minimize social contact. That means being prepared to go awhile without going to the store.  A necessary implication is that everybody’s household supply of TP is inadequate. If they typically run a 10 day supply on average, they need to temporarily increase purchases to actually create what they haven’t had before: a buffer stock.  So people started buying more. Rationally buying more.

Buying a buffer stock is only a temporary increase in demand.  Once you have your buffer, you go back to buying just enough to match usage, keeping the buffer stock in place. You might think that having households add some buffer stock shouldn’t be a big deal, but it was – because that’s how fast TP inventory turns over.

Initially the guidance late last February was for households most at risk to prepare (buy buffer stock) and two weeks was suggested.  But the pandemic has spread faster than most people expected. We now know from Italy’s experience and China’s that extremely broad based and possibly longer  isolation  – meaning minimizing trips to store – is needed. So it’s been an unprecedented and unexpected bump up in purchasing.

But it’s temporary. It  may take weeks for the production chain to catch up, but  purchasing will slow down too. Mills will ratchet up production some.  Nobody is going to truly hoard TP. Contrary to Justin Wolfers’ analysis, TP isn’t like cash. People can keep hoarding cash without limit. TP has some physical limits. Remember it’s high bulk/low value. Where would they put it all?

The dynamics on some other products is a little different. But even with something like hand sanitizer, where the shortages reflect a large sudden increase in consumption and demand, I expect part of the sudden shortage is because we generally have a high-turnover product distribution system. In hand sanitizer, eliminating the shortages depends largely on the ability of manufacturers to switch production facilities away from other products and into more sanitizer production. I expect that to be feasible, but I don’t really know that industry well enough to predict how soon it will restore equilibrium.

There are products in an apparent irrational short supply due to hoarding/panic behavior. I attribute a lot of that to Americans’ fascination with zombie apocalypse movies.

But toilet paper?  It’s a temporary shock to a stable supply chain.

SO relax. Wash your hands and stop being so nasty and judgemental of people.   If you’re going to criticize people, try criticizing them for standing so close in line or not covering their cough.

 

 

GDP 4th Quarter 2010

The numbers for the flash estimate are out today from BEA for U.S. GDP growth, 4th quarter 2010.  I’ll let Calculated Risk summarize the headlines:

Advance Report: Real Annualized GDP Grew at 3.2% in Q4

by CalculatedRisk on 1/28/2011 08:30:00 AM

From the BEA:

Real gross domestic product — the output of goods and services produced by labor and property located in the United States — increased at an annual rate of 3.2 percent in the fourth quarter of 2010, (that is, from the third quarter to the fourth quarter), according to the “advance” estimate released by the Bureau of Economic Analysis.

GDP Growth Rateby CalculatedRisk on 1/28/2011 08:30:00 AM Click on graph for larger image in graph gallery.

This graph shows the quarterly GDP growth (at an annual rate) for the last 30 years. The dashed line is the median growth rate of 3.05%. Growth in Q4 at 3.2% annualized was slightly above trend growth – weak for a recovery, especially with all the slack in the system.

A few observations:

  1. The non-recovery recovery continues.  As I explained earlier, economists haven’t properly defined “recovery”. We define a recession as reduction in GDP (actually a bit more complicated, but close). But we use “recovery” to describe any period after a recession ends.  It’s kind of like a man jumps of a tall building and once he hits the ground and stops falling, we say he must be “recovering” since he isn’t falling.  What appears to be happening is that we are resuming a “normal” growth rate (what’s normal for the last 30 years), but we aren’t really recovering from the recession. Instead, it’s probably better to describe things as those who survived the recession are continuing as they were before, but the 10 million or so unemployed and closed businesses are just out of luck.  They’re out of the picture and we’re moving on without them.
  2. GDP growth in 4th quarter would have been stronger except that businesses had a significant draw-down in inventory.  The inventory draw down surprised me a bit, I’m not sure what it indicates.  We’ll have to wait for the revision to maybe get a better idea. It could be a harbinger of better growth in 1st quarter if businesses seek to replenish that inventory.
  3. We’re still underachieving. It’s gonna be a long time getting back. Technically we have only now gotten real GDP back to a level it was at the end of 2007, the beginning of the recession. But, of course, the economy should have been growing at 2.5-3% per year for these last 3 years.  So the gap between what we are actually producing and what we should be capable of producing continues to be large – on the order of 7-8% of potential GDP. Gee, maybe that’s what those millions of unemployed workers could be used to produce!

One Quick Thing: GDP Not Good

I know I said I’d be on vacation, but thanks to 3G coverage, I can add this item today.  The second revision of 1st qtr GDP 2010 GDP growth rate was announced.  It’s revised down again, and it still looks like it’s mostly all the result of inventory accumulation.  So we have GDP = C + planned Investment + unplanned Inventory accumulation + G + net exports.   And the major reason the overall number is up 2.7% is due to unplanned Inventory accumulation.  We need C and planned Investment to increase — that’s a healthy economy.  But it wasn’t happening in 1st qtr and it looks like things have slipped or slowed since 1st quarter.   So I’m raising my estimate of the chances of a double-dip recession with negative GDP growth in the second half of 2010 to 50% or maybe 60%.

See Calculated Risk:

The Q1 real GDP rate was revised down again (third estimate) to 2.7% from the 2nd estimate of 3.0%.

Consumer spending was weaker in Q1 than originally estimated. PCE growth (personal consumption expenditures) was revised down to 3.0% in Q1 from the previous estimate of 3.5%.

Some more from Reuters: Economy Grew Slower in First Quarter than Expected, Up 2.7%

… business spending, which only rose at a 2.2 percent rate instead of 3.1 percent as reported last month. This was as a spending on structures was revised down to show a slightly bigger decline than reported last month. Growth in software and equipment investment was also lowered to a 11.4 percent rate from 12.7 percent.

Another drag on growth came from exports whose growth was eclipsed by a rise in imports, resulting in a trade deficit that subtracted from GDP.

… real final sales to domestic purchasers, considered a better measure of domestic demand, rose at a 1.6 percent rate instead of the 2.0 percent pace reported last month.

The “Change in private inventories” was revised up to a contribution of 1.88% from the previous estimate of 1.65%. So inventory adjustment accounted for over two-thirds of the GDP growth in Q1 – and the inventory adjustment appears over. This is a weak third estimate.

Posted by CalculatedRisk on 6/25/2010 08:32:00 AM

The Case of the “Disappearing Stimulus” and the “Illusory Inventory”

Worthwhile Canadian Initiative notes that sometimes, a reported slower GDP growth rate is actually better news than another slightly higher reported rate.  As I’ve noted repeatedly on this blog and in class, it’s important to look at the numbers behind the numbers.

When 5.0% GDP growth is better news than 5.9% GDP growth

In 2009Q4, US GDP grew by 5.9% at annual rates; the number was 5.0% in Canada. But our news was much better. Here is a graph of the contributions to GDP growth by expenditure category:

Gdp_contributions_09q4
US GDP growth would have been only 2.0% without the contribution of the inventory terms (which was itself a deceleration in the rate at which stocks were being drawn down.) In Canada, the 2009Q4 GDP number would have been 5.8%.

And look at the contribution of government spending. In the US, the contribution was negative: the increase in federal spending was more than compensated by cutbacks at the state level.

It’s easy to see why the Canadian numbers were greeted with more enthusiasm than were those in the US, even though the headline number was smaller. Growth was evenly distributed across all types of expenditures, and we can expect inventories to bounce back as well fairly soon.

Two observations that I’ll emphasize here.  The first I’ve made before. Most of the reported GDP growth the last 2 quarters of 2009 has actually been inventory-driven.  That’s not sustainable, so we need to see Personal Consumption and Investment growing.  Until they do, the “recovery” is in jeopardy of double-dipping.

The other observation here is that the contribution of government spending in the US was negative.  I’m sure some readers will wonder “How can that be? What about the giant $780 some billion stimulus bill?”. Well the data don’t lie in this case.  We really have NOT had much government stimulus in 2009 in the US for 3 reasons.  Yes, there was a stimulus bill, but a very large portion of it was tax cut, not spending increase.  And as we talk about in class, while a tax cut can be stimulating, it is often not as stimulating as an equal amount of spending because people save part of the tax cut and do not spend all of it. The second reason is because the “giant stimulus” bill was spread out over 2.5 years, not all in 2009.  Finally, while the federal government increased spending, this increased spending has been largely offset by cuts in spending at the local and state levels.  Net impact: negative.    Does this mean that the stimulus bill was a bad idea? that it didn’t work?  NO.  It worked – just imagine how bad it would have been (how even larger the negative impact would have been) if states and local govts had cut and not been offset by the feds.

Inventories & GDP Explained

A good explanation of the role of inventories (and inventory changes) in GDP, including examples from 2009. From Calculated Risk.

First, GDP is Gross Domestic Production. What is being estimated is “domestic production”, but what is being measured is mostly domestic consumption.

Right away we can see that if something is produced domestically and then exported, it will not show up in domestic sales. So exports are added to the equation, and imports subtracted. Investment and Government spending are also added to measures of consumption, and we frequently see an equation like this for GDP:

Y = C + I + G + NX
Y: GDP
C: Consumption
I: Investment
G: Government spending
NX: Exports – imports.

But what about changes in inventories? The same ideas apply. What is measured are sales and changes in inventory, and then production is calculated:

Production = Sales + Changes in Inventory
The following simple table shows how this works, and how it impacts GDP.

Continue reading