As the markets cratered over the past few weeks, the one message I’ve been giving to anyone who would listen is: DO NOT SELL.

I certainly haven’t sold any of my shares. Why would I?

If I was happy to buy something at a price of “x”, I should be even happier to **get 30% more of it for the same price**.

Alas, human nature has a habit of playing nasty tricks on us. Nothing like some cold, hard data to call our lizard brains to order.

But assuming you’ve managed to hold your nerve so far, let’s take the argument above one step further and ask ourselves:

**Is It Time To Buy?**

There are quite a few people out there running complex models and extensive scenario analysis in order to answer this question.

Unfortunately, complex multivariate analysis has major drawbacks.

Firstly, an increase in the number of input variables results in an exponentially larger number of possible permutations.

Secondly, the variables themselves may be correlated, further complicating the analysis.

Finally, in real life, the variables aren’t static. They change all the time, rendering the analysis stale just days, if not hours after it’s been performed.

All of this results in a phenomenon that we investment bankers like call “garbage in – garbage out”. This is the key reason why I refuse to read yet another ten-page research report predicting where the stock market will bottom out.

**Looking To The Past**

One way to dumb it down simplify things could be to look at the outcomes of past bear markets.

As it happens, there are plenty of them, including a ** really nasty one** back in 1929. Could things get even dodgier? Unlikely, though anything is possible.

That being said, the 19 bear markets we’ve had over the past 90 years represent a good starting point – and a meaningful sample size.

The historical bear markets have come in all shapes and sizes. They lasted anywhere between three months and three years.

The declines ranged from 20% (which is what it takes to qualify as a bear market in the first place) to 86% in the ** nasty one**.

Once the markets bottomed out, in some cases it only took them three months to reach their previous all-time highs. In other situations, it took anywhere between one and four years.

And in the two really bad ones, it took as long as 6 years (March 1937) and 13 years (yes, the nasty one back in 1929).

So what does the history tell us?

**Living In A House Of Pain**

The first answer is that we are probably just starting out.

Sure, there have been some bear markets (i.e. Black Monday back in 1987) that ended in just a few months. Blink and you almost missed it!

However, that simply doesn’t seem to be the case here. Given the uncertainty of the situation and the multitude of secondary and tertiary effects, it’s likely the Covid bear market will be with us for a while still.

Which in turn implies that we haven’t yet seen the bottom.

So how much further down will we go before we start digging ourselves out of the hole? In the chart below, I’ve shown where the S&P 500 could bottom out if we follow the pattern of historical bear markets:

The Great Depression resulted in an 86% peak-to-trough decline. Given we are already ~32% down, if things got just as bad as they did 90 years ago we would still have another 54% to go, **all the way to 467.**

The dot-com bear market in 2000 – 2002 saw a 49% decline. That would be another 17% down – **leaving the S&P 500 at ~1,723**.

The Great Recession? Bring on S&P at 1,463 levels.

Ignoring the Great Depression and the bear markets that weren’t as deep, the range of outcomes suggested by the previous recessions is somewhere **between 1,500 and 2,200**.

That’s somewhere between 5% and 65% down from the current levels.

But having said that, is there a silver lining? And the answer is… **a** **resounding yes. **

**Money, Money, Money**

Let’s say you have some spare cash, decide to buck the trend and plonk it into the stock market at some point over the coming months.

It may seem like a crazy move now, but don’t forget the two unifying features of **all bear markets:**

- Sooner or later,
**they end** - At some point after they end, the markets
**recover**–**and exceed**their previous highs.**Always**.

So have a look at the table below, where I show an annualized return on your money in a range of scenarios.

On the left-hand side, I am showing the absolute decline of the market **at the point you decide to go in**, guns blazing. The starting point is 32%, which is where we were as of Friday’s close.

Let’s say you hold your nerve for a while longer – until the market tanks 50%. At that point, the S&P is down to about 1,700 from its peak of ~3,400.

In other words, you need it to go up a full 100% to get back up to where it was in February. That’s the **second column of numbers** on the left-hand side.

Finally, across the top is the time it takes for the market to recover its all-time highs. I’ve picked a range of 1-5 years and but I’m also showing 10 years for all those who believe we are in the Second Great Depression (for the record, I don’t).

Extending the argument above, if you go in after a 50% decline and the market takes 4 years to recover, **you are sitting on a 19% annualized gain** (the numbers shaded in orange).

What if it actually takes 10 years? Sucks to be you – but then again, a 7% nominal return is nothing to laugh at.

Now assume things don’t get much worse than they are today and the 32% haircut to our portfolios is as bad as it gets.

In a three-year recovery scenario, we are looking at a 14% annualized gain. Extend it out to a decade – and you’ve landed at 4%. Not astounding, but sure much better than sitting in bonds.

The analysis above is imperfect. It doesn’t account for the incremental time the stock market will take to bottom out after you invest.

Then again, it also doesn’t account for the fact that after the market recovers its losses, it will keep marching upwards and reach new highs.

I can’t tell you what the future holds. **No one can**. You’ll need to form your own view on whether the correction is “enough” and how long the recovery will take.

But I what I will tell you is that I will be keeping an eye on the table above – and the calendar. Because if there is a silver lining in this situation it is that on April 6^{th }my wife and I can open up new ISAs and put some more money to work.

It may be a rocky ride, but the prize is worth holding on for.

Hi,

Can you explain. I’m not sure I understand the table. Are you saying if for example you do nothing and it takes 5 years to get back to previous peak. This equates to a growth rate of 8%?

Or conversely if the market goes back up 8% a year it will take 5 years for your pot to recover?

In the case the market drops even further then the time to recover is obviously much longer.

This virus style(in my opinion over reacted) self induced crash is more akin to the Great Depression. Businesses basically shut down and potentially bankrupted.

This is not like the 2008 which was just a financial issue. Demand and people were still there.

In that case it does seem more likely than not that the downturn / recession will last much longer and will have much further to fall!

In which case being more cash rich does make more sense!

Sure – let’s use the first line in the table as an example.

The S&P 500 is currently at 2,300. It’s a 32% decline from it’s all-time high of 3,300. To get back to that level, it needs to go up by 47% (the second column).

Now, we know that the market will go back up to that level at some point, but we don’t know when. And yes, there’s a probability of further decline before it starts going up.

Assuming it takes 3 years to get back to 3,300, anyone who invests today will make an annualized 14% return on their money.

If it takes 5 years, it’s an 8% annualized return as you point out above. And if you think we are in a Great Depression situation and it will take 10 years to recover, the implied annualized return is 4%.

Hopefully that’s clearer!

On your second point (comparing covid to the financial crisis) – who knows. Deleveraging shocks typically take longer to play out than demand-side recessions, but it’s really anyone’s guess at this point.

Thanks for the great post. Wouldn’t the S&P go all the way down to 468 in this scenario? An 86.2% decline from the S&P 500 high of 3,393 is much lower than 1,055.

“The Great Depression resulted in an 86% peak-to-trough decline. Given we are already ~32% down, if things got just as bad as they did 90 years ago we would still have another 54% to go, all the way to 1,055.“

You are spot on – I had a mislink in my excel file which I didn’t pick up. Thanks for noticing! I’ve now updated above.

If earnings drops by say 20%, 40% and 60% and PE ratios drop to 15, how level would the S&P500 be at (and the associated decline in %)?

Good question. Let’s use 2021 earnings as 2020 forecasts are a mess at the moment.

For 2021, the current consensus stands at $177 now. With the S&P500 at 2,789 the implied 2021 P/E ratio is 15.75.

A downwards earnings revision by 20% implies $142 of earnings. $142 x 15 = 2,130. That’s roughly 24% off the current levels, with 20% related to the earnings decline and 4% due to multiple contraction from 15.75 to 15.

Hope this helps!