RECENT PROGRESS IN UNDERSTANDING THE ORIGIN OF THE HAWAIIAN-EMPEROR BEND
Recent work invalidates prior studies that inferred significant rates of motion between hotspots since the time of the HEB. Nominal rates of motion are 2 to 6 mm/yr with a lower bound of zero and upper bounds of 8−13 mm/yr (95% c. l.) [Koivisto et al., 2014]. In this context, Hawaiian plume drift as great as 40 mm/yr before 50 Ma B.P. seems unlikely. Other recent work demonstrates the viability of using the orientation of seismic anisotropy in the upper mantle, combined with relative plate motions, to estimate absolute plate motions independently of hotspot tracks. Wang et al.  show that the two reference frames agree with each other within their 95% confidence limits, thus lending credibility to both estimates.
To infer motion of the Hawaiian hotspot relative to the mantle from paleomagnetic data one must ignore true polar wander (TPW), but TPW is too big to ignore and is occurring today—it is an important part of explaining the apparent polar wander (APW) of the Pacific and other plates. New evidence shows that the Hawaiian hotspot was fixed in latitude during formation of most, if not all, of the Emperor seamount chain [Seidman et al., 2017], in contradiction to the southward motion found by Tarduno et al. .
Revised timing and age-dating of the HEB (now ~50 Ma) implies that the change in plate motion coincides with a change in Pacific-Farallon motion and other circum-Pacific tectonic events.
Barkhausen et al  show that the Pacific-Farallon spreading rate doubles between ~50 Ma and ~40 Ma coincident with the acceleration of the Pacific plate from the HEB to the Hawaiian trend and an increasing propagation rate along that trend.
We conclude that current evidence still favors W. J. Morgan’s original explanation for the HEB: that it records a change in Pacific plate motion relative to the deep mantle.