The 2018 market in a 250 year context. Jon Danielsson.

The 2018 market in a 250 year context

December 31, 2018
The financial markets did not have a good 2018 as the media kept on reminding us:

Explanations abound. The president thinks it's the fault of the central bank “Trump Blasts Fed”, while others blame the president "The Trump Volatility Risks".

So, after the markets finally closed, how bad was it? The world’s most important stock index, the S&P 500 fell by -6.4% in 2018 if calculated by log returns. If adjusted by inflation, even bigger at -9%.

The Nikkei 225 did worse,-12.9%, followed by the FTSE at -14.6% while DAX beat them all with -20.7%.

In a historical context, was it as bad as the media would have it? While the S&P 500 index only dates back to 1957, has been reconstructed back to the late 1700s.

I don’t have the other indices that far back, so can only show them for the past 30 years.

For the record, in what follows all prices have been adjusted for inflation, but not corrected for dividends.


By plotting the S&P 500 returns, 2018 does not look all that bad, and there are many worse years.

It is easier to compare 2018 to other years if I sort the returns and identify 2018 in red:

Which years are worse?

Fairly innocuous, 47 years are worse than 2018, making it the 21% worst year in the sample, or about the typical worst year every five years.


2018 was also said to be volatile. If I calculate (realized) volatility:

and as above, identify 2018 in red.

Not very volatile. 48 years have higher volatility than 2018, making it the 22% worst year in the sample, or about the typical worst year every five years.

Recent history and more indices

I don’t have that long a history for many countries. Suppose I use the past 30 years for the US, UK, Japan, and Germany, and calculate how often we can expect to get a worse year than 2018.

Japan, not surprisingly, had a fairly typical year, (it has a lot of bad market years), but for the rest, 2018 was about almost a once a decade event. That is only half the frequency of the full sample, showing how well markets are performed in more recent history.

What about decades?

With all that history, I can also calculate market performance over decades, for example for 2009 to 2018.

We must have lived in good times! Even counting for 2018, the past decade is the fourth best in history.

However, investors can expect to lose more than half their money in 30% of the decades. Proving that the stock market is a dangerous short-term investment.

Is December particularly dangerous?

The worst month of the year was December, and the newspaper headlines liked to tell us that it was the worst December since the Great Depression. Typically, December isn't all that bad, the second safest month after April. In more recent history, it is even better.

December 2018 is the third worst December in history, only beat by 1931 and 1851,

and overall, the 44th worst month, or unconditionally 1.6% worst.

So, while 2018 wasn't all that bad, December was.

So what can we expect?

Unconditionally, we have 102 negative years and 123 positive years, or 55%.

Of the 101 negative years before 2018, 59 are positive and 42 negative, so 58% of the years after a negative year are positive. Slightly better than in a typical year.


At the end of the day, 2018 was an attack by a toothless dragon.

Short and long-term risk
All about BoB — The Bank of England Bot

Models and risk
Bloggs and appendices on artificial intelligence, financial crises, systemic risk, financial risk, models, regulations, financial policy, cryptocurrencies and related topics
© All rights reserved, Jon Danielsson,