SPY:TLT ratio suggests a pending correction

In February we noticed some very short term negative divergence between the SPY and TLT (the 20+ years treasury bonds ETF ): see here. This short term divergence was negated (divergence is only divergence till it isn’t), as stock markets moved higher.

However, large scale negative divergences often foretell of a larger pending correction (see Figure 1).  Currently there’s clear negative divergence between late-February and now: stock markets are up, but also TLT. In fact TLT is up more, hence the SPY:TLT ratio is lower now than it was back then: investors are seeking the safety of treasuries, despite (marginal) higher stock prices.

Figure 1: SPY:TLT ratio over the past 10 years.


The last time, such a scale of divergence occurred was during Q2 of 2016 (April-Jul), and the S&P500 then dropped ~130p before rallying again. Since there’s no negative divergence on the time frame as in 2007, 2011, 2014 and 2015 (several Quarters long), we don’t expect the next correction the SPY:TLT ratio is foretelling us to be of those magnitudes (multi-month), but instead be off the prior magnitude or less.

Given that we still anticipate a rally to ~SPX2500 before the next larger correction ensues, it seems a health dose of selling and fear is needed to allow for the next ~150-200p rally.

3 thoughts on “SPY:TLT ratio suggests a pending correction

  1. Thanks Soul. I’ve been looking into this SPY/TLT ratio since you mentioned it. Fascinating stuff. If we can turn down next week from this bounce to SPX 2389 on Friday, it looks like the ratio analysis would be spot on. Next week will be interesting . Reviewing the most recent drop in March from the SPX 2400 level, it appears the current correction will be similar although in a shorter, more rapid timeframe. Specifically, the a, b, and c waves in early March each took about 1 week to complete leading to a 3 week overall correction. If the correction from 2406 is underway, it appears the a, b waves each took only 1-2 days to complete suggesting that if c down has begun at the end of Friday (or maybe on Monday of the coming week), it should be a fairly rapid decline to 2315-2330 initially and then targeting 2285 or so for completion. Maybe only takes a week or two. Let’s see. Thanks for the spot on analysis.

  2. Pingback: Is the S&P500 heading for 2500 next? | INTELLIGENT INVESTING

Leave a Reply