So I've dug a bit into how a 'normal morphological' form is defined and characterised, since assessing morphology into 'normal' and 'not normal' is always a very tricky thing.
Looking into the method section of the source you cited, we get:
All laboratories generating the data analysed here used standardized
methods for semen analysis [...] The various editions of the manual provided similar
methods for assessing sperm concentration, motility and morphology but
provided different criteria for categorising morphology.
So the problem begins already with the fact that there is not one, but multiple different recommended practices.
(For anyone interested the different staining methods used were: Bryan–Leishman [B], DiffQuik [D], Haematoxylin and Eosin [H], Papanicolaou [P], Quickdip [Q] and Shorr [S]. Of thses H, S and P were used for generation of the 'normal morphology' reference set).
Clearly this problem was also obvious to the authors, so they tried to restrict the data they used:
Although all centres reported using WHO procedures, the recommended
methodologies have changed over time, and many centres have
experienced difficulties with the subjective assessments of
morphology. Data on normal sperm morphology were only included if
results were reported as determined according to the so-called
‘strict’ (Tygerberg) method (WHO, 1992, 1999).
Fortunately these WHO guidelines are available on google books and they require the following points to met (most of which were defined by 95% confidence intervals of control samples):
- head length 4.0-5.0µm
- head width 2.5-3.5µm
- head length-to-width ratio 1.5-1.75
- well defined microsomal region compromising 40-70% of the head area
- mid piece should be slender, <1µm
- [mid piece should be] 1.5x size of the head
- [mid piece should be] axially attached to the head
- Cytoplasmic droplets should be less than half size of the head
- Tail should be [...] approximately 45µm long
Plus one very important sentence:
This classification scheme requires that all 'borderline' forms be considered abnormal
This sentence is important, because it means that each criterium stated has to be fulfilled for a single sperm to classified as morphological normal.
If we take only the numeric requirements (which should cover ~95% of the normal value distribution for a healthy population) and assume that they are not correlated, this means that for any given single healthy sperm we have a $0.95^9 = 0.63 = 63\%$ chance that it is within the 95% interval of all defined parameters.
This means that applying this strict categorisation scheme should even for a 100% healthy population only result around 63% sperm cells categorised as morphologically normal (ignoring that the parameters are probably not interdependent which would make it a bit better).
If we now take into account that the analysed samples were probably not 100% healthy cells and remember that the (reported) difficulties of assessing morphology, I think that the reported values for 'morphological normal' are not as quite as unexpected as they might seem.
Note: all bold highlighting in the quote blocks was added by me.