Proxy data might give a rough idea of the temperature of a century, but is it precise enough to show climate change within a century? And why would we use that when we have much more reliable measurements for the last century?
You do realise that by asking 'why would we use that' is the same as asking why the researchers that built the reconstruction in question should have ever bothered doing so. You are burying your head in the sand as badly as those denying man can affect the climate at all.
The importance of reconstruction covering the instrumental record is to give context to our current warming. We know the planet has been warming for the last century because of our CO2 emissions. Putting that into a context of how normal or abnormal that trend is historically helps us understand the scope of the problem we are creating. Halting the proxy record when the instrumental record begins limits that understanding. The best test of the sensitivity of the proxy sources to current change is to run compare the proxies over the current century as well and compare the result to the instrumental record. The only efforts to that affect I've seen have been in the calibration phase of proxy reconstructions and they have shown a systematic underestimation of recent warming. Identifying the degree of that bias MATTERS to more accurately understanding things.