I recall seeing on this board, a graph of price vs time (recent 2011 time) of VTI/VTSMX (US broad stock index) vs TLT (20-30yr US Treasury bond index), which visually showed their inverse correlation (at least recently).
This led me to wonder if "traditional" "Boglehead-ish" 60% stock/40% bond portfolio, should use TLT for their bond allocation
I checked MadMoneyMachine's http://madmoneymachine.com/professional-lazy-portfolios , & scanned the MMM's list of Boglehead-ish asset allocation (AA) models for the bond asset types used.
I compared the correlations of VTSMX vs these bond funds, using assetcorrelation.com, which uses the div-reinvested AdjClose price from Yahoo Finance.
note: ST - short term, IT - intermediate term, LT- long term
Ticker, Expense Ratio, Asset Type, Avg Maturity, Avg Duration, Yield on 2011-Jun-30, Price history (oldest complete year from Jan-01)
VTSMX, 0.18% or 0.07% for VTI, US Broad stock index, NA, NA, 1.71%, 1997
VTWIX, 0.45% or 0.25% for VT , Global stock index, 1.92%, 2009
-vs.-
TLT, US Treasury 20-30yr index, 0.15%, 28.18yr, 15.93 yr, 4.17%, 2003
EDV, US Treasury 20-30yr Strips index, 0.13% 25.3yr, 27.7yr, 4.20%, 2009
IEF, US Treasury 7-10yr index, 0.15%, 8.63yr, 7.39yr, 2.92%, 2003
PFUIX, 0.50%, ex-US bond active, 6.03yr, 5.04yr, 6.16%, 2005
VBISX, 0.20% or 0.11% for BSV, US ST Bond index, 2.7yr, 2.6yr, 1.98%, 1997
VBMFX, 0.20% or 0.11% for BND, US IT Bond index, 7.4yr, 5.2yr, 3.23%, 1991
VFSTX, 0.22%, US ST Investment Grade Bond active, 3.0yr, 2.3yr, 3.05%, 1991
VIPSX, 0.22%, US Treasury TIPS active, 9.0yr, 4.5yr, 3.51%, 2001
VMLTX, 0.20%, US ST Municipal active, 2.6yr, 2.3yr, 2.29%, 1997
VWEHX, 0.25% US High Yield Corporate Bond active, 5.8yr, 4.8yr, 7.18%, 1990
SHY, 0.15%, US Treasury 1-3yr index, 1.85yr, 1.82yr, 0.94%, 2003
For the period 2004-05-10 to 2011-08-10
VTSMX correlations:
-.40 TLT IEF SHY, -.31 VBISX, -.28 VBMFX, -.26 VFSTX, -.22 VIPSX, -.04 VMLTX, -.08 PFUIX, .20 VWEHX
My interpretation: The best inverse correlation is with the 3 US Treasury indices. Of these perhaps TLT is the best choice for the bond allocation, since the yield is better. Bogleheads could benefit from using TLT. Alternatively, Bogleheads could use SHY, or an equal TLT/SHY mix.
For the more limited recent period of 2008-06-27 to 2011-08-10, I added more recently created funds EDV & VTWIX
VTSMX correlations:
-.45 SHY, -.44 TLT, -.43 IEF, -.39 EDV, -.37 VBISX, -.34 VBMFX, -.28 VFSTX, -.20 VIPSX, -.02 VMLTX, -.05 PFUIX, .22 VWEHX
VTWIX correlations: the bond funds were ranked in the same order as the VTSMX correlations.
My interpretation: same as before (that of the prior 2004-05-10 to 2011-08-10 time frame correlations).
One benefit to a Boglehead AA investor using TLT, say 60% VTSMX 40% TLT, is that if she wanted to switch to a Perm Portfolio approach, it would be easier than it would be for an investor that random bond asset types such as US Muni bonds.
Am I missing something obvious here? Why don't the "experts" that recommend Boglehead-ish AAs, neglect TLT? Perhaps they have a "US in recent years 1980-now", "anchoring bias" (c) MediumTex, & are neglecting the possibility of deflation ala 1989-2011 Japan, where the TLT asset type outperforms all other asset types?
fw: should "traditional" stock/bond portfolios use TLT for bond allocation?
Moderator: Global Moderator