It wasn't worth it.^1 Simply put, anywhere not already conquered was either too well-defended to be worth the bother (the Parthian Empire), or too distant or poor to render suitable booty (Germania, Caledonia, Hibernia). As modern states have realized, invasions were incredibly expensive. One of the reasons for Rome's quick expansion was the operation of a rolling economy, primarily based on slaves.^2 Invasions were paid for by prisoners and material wealth. Showy victories were rewarded by triumphs in Rome and added political capital. Once these two motivations, wealth and glory, became redundant, the Empire became much more defensive than in the glory days of the late Republic and early Emperors. This is an answer which I think is largely satisfactory, but I'll discuss the subject a bit more to flesh it all out.

Gaul is a good example of a wealthy province with the potential for political capital. The Celts were far more sophisticated and wealthy than the Romans often gave them credit.^3 Society was organised into fairly large tribal groupings and, unlike the inhabitants of eastern Germania and Scandinavia, the Celts were settled and very good at agriculture. Some lived in towns and some of these towns were rather large - in the thousands, if not tens of thousands. A centralized, town-based society was easier to militarily dominate than a nomadic, war-like civilization like the Scythians or anyone north and east of the Danube. Gaul was also very wealthy, both in terms of timber and food, but also mineral wealth. It would take the Ottonians in the tenth and eleventh centuries to unlock the mineral wealth of northern Germany. All of these factors made Gaul juicy and ripe for any general with the vision and balls to take it. Roman expansion usually followed along these lines.

Nevertheless, the mechanics of expansion become clearer when we take a closer look at failed invasions (Caledonia and Mesopotamia) and precarious possessions (Britain and Dacia). I won't speak much about Dacia or Mesopotamia, as I know a lot more about Britain, but the latter is worth a comment. The land between the Euphrates and Tigris was never really accessible for Rome; despite its wealth, it happened to be at the centre of the Parthian and then Sassanid Empire, the latter of which was arguably more centralised and efficient than Rome. Armenia was within this disputed area, and regularly switched hands between Rome and the power to the east. Considering Rome's problems elsewhere, it was highly unlikely that the Empire could find the wealth and effort to do an Alexander, so any gains were quickly reversed.^4

As well as taking juicy Gaul, Caesar also attempted a failed invasion of Britain, as I'm sure you know. This was largely to stifle support across the Channel for Caesar's enemies on the Continent (primarily the Belgae). However, Britain was a fabled land of immense wealth, and it wasn't long before an Emperor had a pop. Rome's domination of Britain, following its conquest, depended initially on a combination of military power and oppressive diplomacy, and later on forced Romanisation of the lowlands.^5 The line between the Roman south and militarily occupied north can clearly be seen in the distribution of Roman-style burials and villas on the one hand and inscribed stones on the other.^6 Rome's early control of Britain, however, was incredibly fickle; Boudicca's rebellion and the constant trouble in the north with the Brigantes are two such examples. The conquering of Britain did provide great political capital for Claudius and was obviously considered wealthy and important enough for further campaigning (i.e., Septimius Severus and Tacitus). However, Britain was at the very limit of Rome's authority and the reasons for this parallel why the Empire failed to extend anywhere else after Trajan's death. For one thing, Romanisation was shallow; as I've mentioned, it was confined to certain areas of southern Britain and entirely absent from the north. Even in the south, Roman-style living was heavily dependent on a powerful local economy, linked into larger Roman economic currents. Looking at pottery (an arduous task) following the collapse of Roman power in Britain clearly demonstrates this dependence on the rest of the Empire. Continental styles drop out swiftly, to be replaced by local styles with a very local circulation. This change mirrors the depopulation and destruction of urban centres and subsequent inhabitation of hill forts and other highly defensive and militarised dwellings.^7 The removal of Roman military and economic support from the centre caused necrosis in the limbs; the further from Rome we go, the clearer this becomes.^8

Of course, this isn't the whole story. We might also argue for the importance of the breakdown in political cohesion of the third century in Rome, as well as the role of Christianity in the conversion of subject peoples to good Roman citizens. It also seems somewhat teleological to suggest that Rome just simply reached its "natural" extent. However, much of what happened in the Roman Empire, and concerning it, can be explained in terms of the military and the economy. Romanisation largely failed in Britain as the economy could not support it.^9 Political collapse quickly followed. It is in the failure of further expansion to speak to these two key pillars of Roman power that you might find a clear answer to your question.

FOOTNOTES

^1 This is the majority view in most popular histories. Nothing controversial here.

^2 But Rome was a wonderful civilization and slaves could be made citizens for good work?! Yes, that's partly true. But Rome was a slave state. Following the Punic Wars and other major conflicts in the Late Republic, most people working the land were slaves directly responsible to a member of the nobility. That this was a problem is demonstrated by the constant servile uprisings (yes, Spartacus) elided so elegantly by the narrative sources. During the Pax Romana the lack of new territorial conquests dried up this supply line of human trafficking. This led to increased legal restrictions on freeing slaves. Joshel, S. R., Slavery in the Roman World (2010), and Bradley, K., Slavery and Rebellion in the Roman World (1989).

^3 Lots of interesting sources on this. Look first at Barry Cunliffe's work, in particular The Ancient Celts (now dated), and Europe between the Oceans, and then at the wonderful The Celtic World, ed. M. J. Green. Definitely worth the price.

^4 Sorry I didn't discuss Dacia, it's linked to Trajan and his column. There's some good books cited in the relevant section of Matyszak, The Enemies of Rome.

^5 The Romanisation of Britain is a heavy topic. I'd recommend beginning with one of the top-rated books on amazon and following the bibliographies. Most of my suggestions are specialist articles.

^6 Woolf, 'The Britons From Romans to Barbarians' in Regna and Gentes. The Relationship Between Late Antique and Early Medieval Peoples and Kingdoms in the Transformation of the Rome World, ed. H.-W. Goetz et. al. (2003), pp. 345-80. Also Charles-Edwards, Wales and the Britons, 250-1064 (now in paperback - buy it!).

^7 See the archaeological evidence in Snyder, An Age of Tyrants: Britain and the Britons, A.D. 400-600 (1998). Otherwise, it's quite a boring book.

^8 Caledonia is one of the clearest examples of "not worth the effort" in Roman history. I've already said a lot about Britain, so I'll leave this last detail. I can discuss it in some detail if you like. Also, read Fraser, From Caledonia to Pictland: Scotland to 795. The New Edinburgh History of Scotland is awesome. For more on Roman collapse, I'd recommend Heather, P., The Fall of the Roman Empire (2006).

^9 This is incredibly contentious. It is slightly naughty of me mentioning it only briefly. These articles support my view, though others certainly disagree: Reece, R., 'The End of the City in Roman Britain' in The City in Late Antiquity, ed. J. Rich (1992), pp. 136-44, and Faulkner, N., 'The Case for the Dark Ages' in Debating Late Antiquity in Britain AD 300-700, British Archaeological Reports, 365, ed. R. Collins and J. Gerrard (2004), pp. 1-12. For another view, see the two books by Ken Dark, which I don't recommend. Faulkner deconstructs Dark's thesis well.