Why Buy Stocks (Especially Ones That Don’t Pay Dividends)?

I recently asked myself “Why do people by stocks”?  It seems like a simple question but I quickly hit a roadblock in my line of thinking.  By the way, when I say “people”, I literally mean everyone, including both large institutional investors and individual investors like you and me.

The content of this article may seem obvious to some but in speaking with a few others I realized that many people, including myself prior to this writing, don’t understand some fundamentals.

My Confusion

My line of thinking went like this: You buy shares in a company because you want a return on your investment (ie. to make more money).  The way in which you make money is either through sharing in the company’s profits (ie. a dividend or stock buyback) or through selling your shares at a higher price (capital gains).  But this latter method depends on the existence of people in the market willing to buy your shares at that higher price.  But why are they willing to do that? They have the same motivations as you.  They are expecting either a share of the company’s profits or, in turn, to sell at a higher price.  Again, this latter point depends on the existence of yet another person willing to buy at that even higher price.  But why is this third person willing to do that?  You can see how this continues in an infinite regression that seems kind of insane.  But I thought this was all fine and good as long as the company eventually shares its profits with the shareholders.  As the shares change hands, each owner is basically making a bet on how big their share of the profit will be (ie. betting on the company’s profitability) when it’s finally paid out.  But we all know many companies don’t directly share their profits with shareholders.  So why are people buying stocks in these companies?  What is their incentive?  Doesn’t this make the trading of these shares look like a Ponzi/pyramid scheme?  (I’m not saying that it’s a Ponzi scheme, just that the infinite regression qualities seem similar).

I tried researching this question and didn’t find much clear information online.  I also spoke with several people and nobody could give me a good explanation.  I realized that a lot of people, including myself, don’t fundamentally understand why a stock’s value fluctuates and why people invest in the first place.  It seems like a lot of people invest simply because they think the value will “go up” because of the company is “doing well” without really understanding what’s happening.  What enables the value to go up?  What does the share price even mean?  How do we clarify the confusion about the infinite regression?

The Basic Answer

The fundamental answer to all of this is that the profits must be shared with shareholders at some point.  This is what terminates the infinite regression.  If a company believes that they can retain their earned profits to further grow and generate even more profits, they will do so.  However if a company continues to grow and do well, eventually they will accumulate so much cash in the bank that they can’t find good use for all of it.  At this point they will have to pay it out to shareholders through a dividend or stock buyback.  If they refuse to, at some point the shareholders will band together and vote for new management that will pay it out.

In summary, the fundamental bet that is made by purchasing stock is that the company’s profits will eventually be paid out to shareholders.  It’s often true that your personal intention may be to sell before that happens and make a return on capital gains.  But the thing that enables you to do that — in other words, the thing that incentivizes the infinite regression of buyers is that eventually the profits will be shared.

Why Do Share Prices Fluctuate?

The share price on a given day is simply what people are willing to buy/sell it for on that day.  How do people decide what they are willing to pay?  A lot of complex analysis goes into this but it can all be summarized as basically trying to guess the company’s future profitability.  If people believe a company will be more profitable in the future and thus pay out more money to shareholders, the stock will be worth more.  In contrast if people believe the company will be less profitable, say because of some bad news or political uncertainty, the share price will decrease.  But as long as people believe some profits will be made the share price will stay above $0.  If everyone believes the company is going to dissolve, the share price will drop to $0 because what’s the point of investing if the company will never make any profits to share with shareholders?

Berkshire Hathaway

An interesting example is Warren Buffett’s company, Berkshire Hathaway, which is one of the most successful companies of all time.  Its book value has grown an average of almost 20% per year for the last 48 years and it has a market cap of $286B today.  At the end of 2012 Berkshire reported it had more than $40B cash in the bank and Retained Earnings of $124B, meaning that since the company’s inception it has re-invested $124B of the profits instead of paying them out [1].  Berkshire famously does not pay dividends despite having lots of cash in the bank.  Why?  Warren Buffett has continuously believed that he can get a positive return on every dollar he has in the bank, he has proven this year after year, and therefore says he will never pay a dividend [2].  This makes sense — he has shown he can get a 20% return on every dollar which is likely a lot better than what investors could otherwise get if they put the money elsewhere.  So why do people buy shares in Berkshire even though there will be no dividends in the foreseeable future?  The belief is that Berkshire’s ever-growing profits will eventually be distributed.  Nobody knows exactly when but it will happen when Berkshire is no longer able to successfully re-invest its profits.  If they don’t pay it out at this point, shareholders will vote to make it so.

I didn’t talk about acquisitions because they are just a special case of investing where you purchase 100% of the shares.  In an acquisition you are still motivated by the same things: you think the future profits justify the investment.

The only other article I found on this subject is [3].  It’s a good read and concludes the same thing: that the shareholders can vote to have dividends paid out.

If you have any feedback on what I’ve written please share it in the comments!

References

[1] http://www.berkshirehathaway.com/2012ar/2012ar.pdf, Page-30
[2] http://youtu.be/aL766NK2ynw?t=2m28s
[3] http://beginnersinvest.about.com/od/dividendsdrips1/a/why-stocks-without-dividends-can-still-be-a-good-investment.htm

How is Your Cable Connection Bidirectional?

When you think of your home’s cable connection, it seems like a unidirectional thing since many TV channels come into your home through a single wire.  But what about when you order a movie on-demand?  Your choice of movie must go out to the cable company somehow, over the same wire.  What about cable internet access?  When you surf the web and send emails you send out tons of data.  Cable home phone is another popular service as well — your voice has to be transmitted out from your home.  And to top it off, many people subscribe to all 3 services at once!

How is it possible to have all this information traveling simultaneously in both directions on one cable?

It’s a simple question with a relatively simple answer.  It was easy to find out how multiple services can exist on one cable unidirectionally but harder to find out how they can exist bidirectionally.

How to Transmit Several Things in One Direction

It helps to talk about the unidirectional case first.  The answer to this first question is that many blocks of information can be sent over one wire using a technique called Frequency Division Multiplexing (FDM) [1].  TV cables have a bandwidth of 1 GHz [2] which means 1 GHz is the maximum frequency they can carry.  FDM allows this 1 GHz spectrum to be split up into slots (say 6 MHz each) where each slot carries a block of information.  In the case of old-school analog cable TV, each slot carried one channel [2].  The receiver, say your TV or set-top-box, uses a filter to tune in to a specific slot to extract its information.

These slots can carry any type of electronic information including digital TV channels, internet data, and voice (phone) data.  The receiver only looks at the slots it’s interested in.  For example, your cable modem simply tunes into the slots that are allocated for internet data and ignores the slots that carry TV channels.

I won’t go into details about how FDM is implemented at the circuit-level but see the Wikipedia links below for more info.

How Bidirectional Works

Bidirectional transport is also called “duplex”.  A simple way to implement duplex communication is to have separate wires for each direction.  This is how Ethernet works for example.  But your home’s cable connection is just a single wire.  So how does it work?

As you may have guessed, FDM is used to allocate separate slots for receive and transmit data.  So this means that both ends of the cable are being driven simultaneously.  Your modem is driving the wire with data to send out to your provider and your provider is also driving the wire with data to send to you!  Intuitively this seemed off to me because how can two things drive the same cable at once without creating interference?  There are 2 parts to the answer:

  1. From an electrical perspective, the principle of superposition allows two transmitters to drive the same cable at once with the resulting signal on the cable simply being the sum of the two individual signals. [3]
  2. By using FDM to first assign these signals into separate slots they can be safely summed together without interfering with each other.  If they were in the same slot the resulting sum would be corrupt since the signals would directly overlap.

The circuit that performs this function is called a diplexer [4].  When the diplexer is used for bidirectional communication, it’s called a duplexer [5].  See [6] for more information on duplexers.

References

  1. http://en.wikipedia.org/wiki/Frequency_division_multiplexing
  2. http://en.wikipedia.org/wiki/Cable_tv#How_it_works
  3. http://en.wikipedia.org/wiki/Superposition_theorem
  4. http://en.wikipedia.org/wiki/Diplexer
  5. http://en.wikipedia.org/wiki/Duplexer
  6. http://www.rfsolutions.com/duplex.htm

Adding Salt to Ice in a Cooler

Why do people sometimes add salt to ice in a cooler for drinks or beer?  It’s a simple question.  There are tons of articles and videos online but I couldn’t find a comprehensive explanation of why it works.

Some Basics

It’s important to first understand what melting point and freezing point mean.

Melting point is a property of solids.  The melting point is simply the temperature at which the solid turns into a liquid.  For example, the melting point of normal ice is 0°C at standard pressure.

Freezing point is a property of liquids.  The freezing point is simply the temperature at which the liquid turns into a solid.  Thus the freezing point of normal freshwater is 0°C.

The value of the melting and freezing points is the same (ie. both are 0°C for water).

So as you can see, melting point and freezing point are kind of “mirrors” of each other.

It doesn’t make sense to talk about the freezing point of a solid (since it’s already solidified).

What Does Salt Do?

When salt is added to water it lowers the freezing point.  In other words the water needs to be chilled to a temperature lower than 0°C (say -2°C for example) in order to change to ice.  Another way to say the same thing is that salt allows water to exist as a liquid at a temperature lower than 0°C.  This is the important part that’s relevant to beer coolers.  We’ll come back to this later.

For completeness let’s look at this from the perspective of melting point.  When salt is added to ice it lowers the melting point.  In other words the ice begins melting at a temperature lower than 0°C.  This is why salt is added to ice on the roads in the winter.  It causes ice, that would have otherwise remained as a solid in sub-zero temperatures, to turn to water.  Note that the temperature of the water has not changed.  It’s still at a sub-zero temperature but, as mentioned above, the salt allows it to remain as a liquid at the lower temperature.  Don’t think that just because the salted-ice has become water the temperature has risen.

How Does this Apply to Beer Coolers?

In two ways: chilling beers fast and keeping beers cold.

Chilling Beers Fast

An important part of chilling beer fast is to maximize how much of the can/bottle’s surface area is in contact with the chilling agent, whether it’s ice, water, or cold air.

Salt-water lets you chill a warm beer really fast — much faster than a freezer [1].  Pour cold water, ice, and salt into the cooler to create a salt-water-ice bath.  Why does it work?  The ice will be at a temperature way below zero, usually -18°C for a household freezer [2].   The ice will cool the water down and the salt will allow the water temperature to drop below 0°C.  The beer will then be fully submersed in sub-zero water, maximizing the surface area in contact.

Without salt the water will remain at a temperature slightly higher than 0°C even though you have -18°C ice cubes floating in it.  The water will still get quite cold and will do a good job but not as cold as salt-water.

If you only put ice in the cooler then less of the beer’s surface is touching the chilling agent (ice) since ice cubes are irregularly-shaped.  It won’t chill the beer as fast as salt-water.

Keeping Beers Cold

If your goal is to keep the beer cold for a long time (ie. if you go camping or on a picnic) it’s still a good idea to add some water and salt to the ice because it will make your beers colder initially which means they’ll stay cold longer.  The salt will make the ice melt as well but the resulting water will still be very cold.  The specific heat capacity of water is double that of ice [3], which means, in theory, the sub-zero water will stay colder longer than plain ice will.  You could argue that the starting temperature of ice is much lower (-18°C) than sub-zero water (which might be just a few degrees below 0°C).  So it’s a bit of a tradeoff.  Also, adding salt to the water does lower its heat capacity but not by much [4].

Note: you need lots of salt (handfuls) for the above stuff to work.  Table salt is fine.

References

  1. http://youtu.be/k3PvQXlxvZY
  2. http://www.inspection.gc.ca/food/consumer-centre/food-safety-tips/food-handling/kitchen/eng/1329169036268/1329169127007
  3. http://en.wikipedia.org/wiki/Properties_of_water#Water.2C_ice_and_vapor
  4. http://www.kayelaby.npl.co.uk/general_physics/2_7/2_7_9.html