In the same way that there is a well-recognized asset-class called “Asia-Pacific ex-Japan”, I have proposed (for my own use) the following fixed-income class: “Multi-Sector ex-High Yield” which includes US government bonds of any maturity and investment-grade corporates of any maturity, but not “bank debt”, “preferreds”, “converts”, “world bonds”, or “emerging market debt” (which are the other sub-categories that Morningstar includes under the heading of “Taxable Bond Funds”.) Furthermore, I consider “Multi-Sector ex-HY” to be merely a souped-up money-market fund whose purpose is cash management, rather than an FI investing objective.

Explanation: The conventional distinction between a MM fund and a short-term bond fund is the implied assurance that principal will not be impaired in a MM fund and the (SEC?) requirement that the portfolio's average maturity be less than 13 months. However, I'm not willing to sacrifice yield for the sake of definitional purity. If a putatively investment-grade bond is in danger of defaulting, ipso facto, it ain't investment grade. Therefore, I set the trigger point between invest-grade and spec-grade higher than is customarily done, and, furthermore, I work with notions of multi-dimensional “thresholds” and “transitions” instead of single-factor, trigger points (such as an S&P or Moody's rating).

Further explanation: If a bond even hints at being risk for default, I treat it as a junk bond, and there is no credit rating that I won't buy, including bonds already in default and slapped with a “D” rating. But I do that buying as a total-returns, high-yield investor (aka, junk bond investor), not as an incautious, but yield-hungry investment-grade bond buyer.

If I want junk, I buy it. But if I want investment-grade bonds, I don't take risks with lower-tier credits. Therefore, I do not look at mid-tier, investment-grade ratings with the same eyes as the average FI investor who makes a careful distinction between credits that would “properly” be part of a money market portfolio and credits that would characterize the holdings of a typical corporate bond fund (of any maturity).

I want either the maximum amount of total return I can get for the risks that I think I can manage (which might include total loss of the position), or I want the maximum total return I can get while having virtual certainty that there will be no impairment of principal. There is no middle ground for me. Such an approach is a bar-bell strategy, or it might be called a Yoda strategy:

“There is 'DO' and 'NOT DO'. There is no 'TRY-TO-DO'.”